CN113379572A

CN113379572A - House source explanation method and device, computer readable storage medium and electronic equipment

Info

Publication number: CN113379572A
Application number: CN202110632765.7A
Authority: CN
Inventors: 彭嵩琪
Original assignee: Beijing Fangjianghu Technology Co Ltd
Current assignee: Seashell Housing Beijing Technology Co Ltd
Priority date: 2021-06-07
Filing date: 2021-06-07
Publication date: 2021-09-10

Abstract

The embodiment of the disclosure discloses a house source explanation method and device, a computer readable storage medium and an electronic device, wherein the method comprises the following steps: acquiring relevant information of a current house source according to a viewing request of a current user to the current house source; wherein, the related information comprises description text, at least one related broker and introduction video; a first lecture video generated based on the lecture text and relevant information of a first broker of the at least one relevant broker; processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video; obtaining a house source video explaining the current house source for the current user based on the second lecture video and the introduction video; the embodiment of the disclosure enhances the participation of the broker, improves the service perception and service experience of the client, enables the explanation of the house resources to better meet the requirements of the current user, and improves the efficiency of acquiring the user information.

Description

House source explanation method and device, computer readable storage medium and electronic equipment

Technical Field

The present disclosure relates to housing technology, and in particular, to a housing source explanation method and apparatus, a computer readable storage medium, and an electronic device.

Background

In the prior art, the video explanation of the house source is usually that a broker records voice to provide a house speaking service for a client, but the product needs a lot of effort of the broker, and the service quality of the broker has large variance, so that the experience of the client cannot be guaranteed; an AI lecturer is also proposed in the prior art, which uses a virtual cartoon image as a lecturer to automatically generate a manuscript and an lecture voice according to house information.

Disclosure of Invention

The present disclosure is proposed to solve the above technical problems. The embodiment of the disclosure provides a house source explanation method and device, a computer readable storage medium and an electronic device.

According to an aspect of an embodiment of the present disclosure, there is provided a house source explanation method including:

acquiring relevant information of a current house source according to a viewing request of a current user to the current house source; wherein, the related information comprises description text, at least one related broker and introduction video;

a first lecture video generated based on the lecture text and relevant information of a first broker of the at least one relevant broker;

processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video;

and obtaining a house source video explaining the current house source for the current user based on the second lecture video and the introduction video.

Optionally, the first lecture video generated based on the lecture text and the related information of the first broker of the at least one related broker comprises:

determining the first broker according to the contribution value of the at least one relevant broker to the current house source;

generating introduction audio corresponding to the tone of the first menstrual person according to at least one preset audio corresponding to the first menstrual person and the description text;

obtaining a first lecture video corresponding to the introduction audio based on the introduction audio and at least one preset action video corresponding to the first person; the first lecture video comprises at least one description part, and the description part comprises at least one lecture data.

Optionally, the obtaining a first lecture video corresponding to the introduction audio based on the introduction audio and at least one preset action video corresponding to the first person includes:

processing the at least one preset action video with the first duration based on a second duration of the introduction audio to obtain a processed video with the second duration;

and replacing mouth movements in the processed video according to the introduction audio to obtain a first lecture video in which the first person speaks the introduction audio.

Optionally, the processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video includes:

obtaining a user representation of the current user;

determining whether the user representation affects a ranking of the point of speech data;

in response to the user representation influencing the sorting of the talk point data, deleting and/or reordering the talk point data of the first lecture video according to the user representation; and obtaining the processed second lecture video.

Optionally, the method further comprises:

determining at least one preset questioning node included in the house source video based on the house source video and the description text;

for the at least one preset questioning node, determining a target question for the preset questioning node according to an extraction rule;

and adjusting the extraction rule according to the feedback result of the current user.

Optionally, the determining a target question for the preset question node according to the extraction rule includes:

determining a ranking of the portrait tags according to a weight value of at least one portrait tag included in the user portrait of the current user;

determining a target portrait label corresponding to the preset questioning node based on the sequence of the portrait labels;

and taking the question corresponding to the target portrait label as the target question.

Optionally, before determining the ranking of the portrait tags according to a weight value of at least one portrait tag included in the user portrait of the current user, further comprising:

classifying at least one problem corresponding to the current house source according to the corresponding portrait label; wherein, the portrait label has a corresponding relation with the question;

performing first screening on the at least one question based on the portrait label corresponding to the question determined by the classification to obtain a first candidate question set; wherein the first set of candidate questions is empty or comprises at least one candidate question;

performing second screening on the first candidate problem set according to the portrait label and the related information of the current house source to obtain a second candidate problem set; wherein the second set of candidate questions is empty or comprises at least one candidate question;

determining a ranking of the portrait tags according to a weight value of at least one portrait tag included in the user portrait of the current user, comprising:

determining an ordering of at least one portrait label corresponding to the second candidate question set based on a weight value of the portrait label corresponding to the candidate question included in the second candidate question set.

Optionally, the performing, on the basis of the portrait label corresponding to the question determined by the classifying, a first filtering on the at least one question to obtain a first candidate question set includes:

determining whether the portrait label corresponding to the problem has corresponding data in the user portrait of the current user according to the classification;

deleting the question from the at least one question when corresponding data exists;

and when the corresponding data does not exist, adding the problem serving as a candidate problem into the first candidate problem set.

Optionally, the method further comprises:

and determining one question template from at least one question template corresponding to the determined target question.

Optionally, the adjusting the extraction rule according to the feedback result of the current user includes:

storing the target question and the feedback result, and determining an answer rate of the question based on the stored feedback result of the at least one question;

and adjusting the weight value of the portrait label corresponding to at least one question based on the answer rate of the question.

Optionally, the feedback result comprises an answer or no answer;

before the storing the target question and the feedback result and determining the answer rate of the question based on the stored feedback result of the at least one question, the method further comprises:

determining whether the current user answers the target question;

responding to the target question answered by the current user, providing a corresponding answer for the current user, supplementing the user portrait, and updating a weight value of corresponding talk point data according to the supplemented user portrait; playing the talk point data according to the updated weight value;

and responding to the target question which is not answered by the current user, waiting for a set time or providing a corresponding answer for the current user, and recording the unanswered state of the current user.

Optionally, the information related to the current house source further includes: at least one bright spot image and bright spot description information corresponding to the bright spot image;

further comprising:

determining a bright point explanation video corresponding to the bright point image based on the bright point description information corresponding to the at least one bright point image and at least one preset action video corresponding to the first busier;

determining a bright point display video based on the bright point explanation video, the bright point image and the bright point description information;

sorting the at least one highlight display video based on the user image of the current user;

and displaying the at least one bright point display video according to the sequence.

Optionally, before obtaining the relevant information of the current house source according to the viewing request of the current user to the current house source, the method further includes:

receiving at least one relevant data corresponding to at least one room source in a database input by at least one terminal;

determining the authenticity of the related data by matching the related data input by at least one terminal corresponding to the at least one house source;

and storing the relevant data with the authenticity determined as real into the relevant information of the corresponding house source in the database.

According to another aspect of the embodiments of the present disclosure, there is provided an origin explaining apparatus including:

the house source information acquisition module is used for acquiring the related information of the current house source according to the viewing request of the current user to the current house source; wherein, the related information comprises description text, at least one related broker and introduction video;

a first lecture video module for generating a first lecture video based on the lecture text and information about a first broker of the at least one relevant broker;

the second lecture video module is used for processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video;

and the house source explanation module is used for obtaining the house source video explaining the current house source for the current user based on the second explanation video and the introduction video.

Optionally, the first lecture video module is specifically configured to determine the first broker according to a contribution value of the at least one relevant broker to the current house source; generating introduction audio corresponding to the tone of the first menstrual person according to at least one preset audio corresponding to the first menstrual person and the description text; obtaining a first lecture video corresponding to the introduction audio based on the introduction audio and at least one preset action video corresponding to the first person; the first lecture video comprises at least one description part, and the description part comprises at least one lecture data.

Optionally, when the first lecture video corresponding to the introduction audio is obtained based on the introduction audio and the at least one preset action video corresponding to the first parent person, the first lecture video module is configured to process the at least one preset action video with the first duration based on a second duration of the introduction audio to obtain a processed video with the second duration; and replacing mouth movements in the processed video according to the introduction audio to obtain a first lecture video in which the first person speaks the introduction audio.

Optionally, the second lecture video module is specifically configured to obtain a user representation of the current user; determining whether the user representation affects a ranking of the point of speech data; in response to the user representation influencing the sorting of the talk point data, deleting and/or reordering the talk point data of the first lecture video according to the user representation; and obtaining the processed second lecture video.

Optionally, the apparatus further comprises:

the node identification module is used for determining at least one preset question node included in the house source video based on the house source video and the description text;

the question extraction module is used for determining a target question for the preset question node according to an extraction rule for the at least one preset question node;

and the rule adjusting module is used for adjusting the extraction rule according to the feedback result of the current user.

Optionally, the question extraction module is specifically configured to determine a ranking of the portrait tags according to a weight value of at least one portrait tag included in the user portrait of the current user; determining a target portrait label corresponding to the preset questioning node based on the sequence of the portrait labels; and taking the question corresponding to the target portrait label as the target question.

Optionally, the question extraction module is further configured to classify at least one question corresponding to the current house source according to the portrait label; wherein, the portrait label has a corresponding relation with the question; performing first screening on the at least one question based on the portrait label corresponding to the question determined by the classification to obtain a first candidate question set; wherein the first set of candidate questions is empty or comprises at least one candidate question; performing second screening on the first candidate problem set according to the portrait label and the related information of the current house source to obtain a second candidate problem set; wherein the second set of candidate questions is empty or comprises at least one candidate question;

the question extraction module is used for determining the ranking of at least one portrait label corresponding to the second candidate question set based on the weight value of the portrait label corresponding to the candidate question included in the second candidate question set when determining the ranking of the portrait label according to the weight value of at least one portrait label included in the user portrait of the current user.

Optionally, when the portrait tag corresponding to the question determined based on the classification performs first screening on the at least one question to obtain a first candidate question set, the question extraction module is configured to determine, according to the classification, whether the portrait tag corresponding to the question has corresponding data in the user portrait of the current user; deleting the question from the at least one question when corresponding data exists; and when the corresponding data does not exist, adding the problem serving as a candidate problem into the first candidate problem set.

Optionally, the question extracting module is further configured to determine one question template from at least one question template corresponding to the determined target question.

Optionally, the rule adjusting module is specifically configured to store the target question and the feedback result, and determine an answer rate of the question based on the stored feedback result of the at least one question; and adjusting the weight value of the portrait label corresponding to at least one question based on the answer rate of the question.

Optionally, the feedback result comprises an answer or no answer;

the rule adjusting module is further used for determining whether the current user answers the target question; responding to the target question answered by the current user, providing a corresponding answer for the current user, supplementing the user portrait, and updating a weight value of corresponding talk point data according to the supplemented user portrait; playing the talk point data according to the updated weight value; and responding to the target question which is not answered by the current user, waiting for a set time or providing a corresponding answer for the current user, and recording the unanswered state of the current user.

the device further comprises:

the bright point video module is used for determining a bright point explanation video corresponding to the bright point image based on the bright point description information corresponding to the at least one bright point image and the at least one preset action video corresponding to the first businessman; determining a bright point display video based on the bright point explanation video, the bright point image and the bright point description information; sorting the at least one highlight display video based on the user image of the current user; and displaying the at least one bright point display video according to the sequence.

Optionally, the apparatus further comprises:

the information acquisition module is used for receiving at least one piece of relevant data corresponding to at least one house source in a database input by at least one terminal; determining the authenticity of the related data by matching the related data input by at least one terminal corresponding to the at least one house source; and storing the relevant data with the authenticity determined as real into the relevant information of the corresponding house source in the database.

According to still another aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium storing a computer program for executing the method for explaining a house source according to any one of the embodiments.

According to still another aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:

a processor;

a memory for storing the processor-executable instructions;

the processor is configured to read the executable instructions from the memory and execute the instructions to implement the house source explanation method according to any one of the above embodiments.

Based on the house source explanation method and device, the computer readable storage medium and the electronic device provided by the above embodiments of the present disclosure, the relevant information of the current house source is obtained according to the current user's request for viewing the current house source; wherein, the related information comprises description text, at least one related broker and introduction video; a first lecture video generated based on the lecture text and relevant information of a first broker of the at least one relevant broker; processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video; obtaining a house source video explaining the current house source for the current user based on the second lecture video and the introduction video; the first person is confirmed through the house source related information, the second lecture video which is described by the first person image is obtained, the participation degree of brokers is enhanced, the service perception and the service experience of customers are improved, the user portrait is combined, the explanation of the house source is more fit with the requirements of the current users, and the efficiency of obtaining user information is improved.

The technical solution of the present disclosure is further described in detail by the accompanying drawings and examples.

Drawings

The above and other objects, features and advantages of the present disclosure will become more apparent by describing in more detail embodiments of the present disclosure with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the principles of the disclosure and not to limit the disclosure. In the drawings, like reference numbers generally represent like parts or steps.

Fig. 1 is a schematic flow chart of a house source explanation method according to an exemplary embodiment of the present disclosure.

Fig. 2 is a schematic flow chart of step 104 in the embodiment shown in fig. 1 of the present disclosure.

Fig. 3 is a schematic flow chart of step 1043 in the embodiment shown in fig. 2 of the present disclosure.

Fig. 4 is a schematic flow chart of step 302 in the embodiment shown in fig. 3 of the present disclosure.

Fig. 5 is a schematic flow chart of step 106 in the embodiment shown in fig. 1 of the present disclosure.

Fig. 6 is a schematic flow chart of obtaining a target question according to an extraction rule in an example of the room source explanation method proposed by the present disclosure.

Fig. 7 is a schematic flowchart illustrating a question interaction with a current user in an example of a room source explanation method proposed by the present disclosure.

Fig. 8a is a schematic diagram of a special effect animation set for the features of the full-clear pattern of the house source in one example of the house source explanation method proposed by the present disclosure.

Fig. 8b is a schematic diagram of a special effect animation set for the north-south permeability feature of the origin in one example of the origin explanation method proposed by the present disclosure.

Fig. 8c is a schematic diagram of a special effect animation set for a characteristic of dynamic and static separation of a room source in an example of a room source explanation method provided by the present disclosure.

Fig. 8d-1 is a schematic diagram of a part of special effect animation set for features of a kitchen and a toilet of a house source in one example of the house source explanation method provided by the present disclosure.

Fig. 8d-2 is another part of the special effect animation diagram set for the characteristics of the Ming kitchen and Ming toilet of the house source in one example of the house source explanation method proposed by the present disclosure.

Fig. 9 is a schematic structural diagram of an atrioventricular source explanation device according to an exemplary embodiment of the present disclosure.

Fig. 10 is a block diagram of an electronic device provided in an exemplary embodiment of the present disclosure.

Detailed Description

Hereinafter, example embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of the embodiments of the present disclosure and not all embodiments of the present disclosure, with the understanding that the present disclosure is not limited to the example embodiments described herein.

It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

It will be understood by those of skill in the art that the terms "first," "second," and the like in the embodiments of the present disclosure are used merely to distinguish one element from another, and are not intended to imply any particular technical meaning, nor is the necessary logical order between them.

It is also understood that in embodiments of the present disclosure, "a plurality" may refer to two or more and "at least one" may refer to one, two or more.

It is also to be understood that any reference to any component, data, or structure in the embodiments of the disclosure, may be generally understood as one or more, unless explicitly defined otherwise or stated otherwise.

In addition, the term "and/or" in the present disclosure is only one kind of association relationship describing an associated object, and means that three kinds of relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" in the present disclosure generally indicates that the former and latter associated objects are in an "or" relationship. The data referred to in this disclosure may include unstructured data, such as text, images, video, etc., as well as structured data.

It should also be understood that the description of the various embodiments of the present disclosure emphasizes the differences between the various embodiments, and the same or similar parts may be referred to each other, so that the descriptions thereof are omitted for brevity.

Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.

The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses.

Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.

The disclosed embodiments may be applied to electronic devices such as terminal devices, computer systems, servers, etc., which are operational with numerous other general purpose or special purpose computing system environments or configurations. Examples of well known terminal devices, computing systems, environments, and/or configurations that may be suitable for use with electronic devices, such as terminal devices, computer systems, servers, and the like, include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, microprocessor-based systems, set top boxes, programmable consumer electronics, network pcs, minicomputer systems, mainframe computer systems, distributed cloud computing environments that include any of the above systems, and the like.

Electronic devices such as terminal devices, computer systems, servers, etc. may be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, etc. that perform particular tasks or implement particular abstract data types. The computer system/server may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.

Summary of the application

In the process of implementing the present disclosure, the inventor finds that in the automatic house-speaking product in the prior art, the virtual cartoon image is taken as the instructor, and at least the following problems exist: the professionalism of the broker can not be reflected, the service perception of temperature can not be transmitted to the client, and the client is easy to be tired due to the uniform image and sound.

Exemplary method

Fig. 1 is a schematic flow chart of a house source explanation method according to an exemplary embodiment of the present disclosure. The embodiment can be applied to an electronic device, as shown in fig. 1, and includes the following steps:

and step 102, acquiring relevant information of the current house source according to a viewing request of the current user to the current house source.

The related information comprises description text, at least one related broker and introduction videos.

In this embodiment, the relevant information of the current house source may be obtained from a database storing a large amount of relevant information of the house source, at least one house source stored in the database corresponds to one piece of data, and the relevant information of the house source is stored in the piece of data; for example: basic information including house resources: broker information, cell corollary (profile, school, traffic, life, etc.), cell internal description (basic information, property/security, cell internal facilities, etc.), house analysis (building information, house basic information, bay band bay information, house type analysis, etc.), trade information (price, trade attributes, owner information, other information, etc.), etc.; based on the information in the database, the description text, the relevant broker and the introduction video of the current house source can be obtained, and in addition, the relevant information can also comprise highlight images and the like.

A first lecture video is generated based on the lecture text and information about a first broker of the at least one relevant broker, step 104.

In an embodiment, a house may correspond to at least one relevant broker, where the relevant broker indicates that the broker has a certain contribution value to the house, and the contribution value may be obtained according to behaviors of the broker such as field investigation, taking a look at the house, filling in relevant information of the house, explaining for a user, and the like (for example, filling in an effective house-related information product 1 score, and the like, and different behaviors may be set for different scores, and a specific score distribution condition may be set according to an actual scene); the related information of the first person may include face information and tone information of the first person, and optionally, the first lecture video may be a video that speaks a lecture text in a tone and a character of the first person.

And 106, processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video.

In this embodiment, the user representation includes information about the user, for example, the user representation includes but is not limited to: user basic information, house purchasing preference information, behavior data information and the like; the embodiment combines the user image to process the first lecture video, improves the matching degree of the obtained second lecture video and the current user, and is more personalized.

And step 108, obtaining a house source video explaining the current house source for the current user based on the second explanation video and the introduction video.

Optionally, the introduction video includes some description pictures of the current house source, and these description pictures correspond to the description text, so that the second lecture video obtained based on the description text can also correspond to the introduction video, and the second lecture video and the introduction video are combined based on the corresponding relationship, so that the house source explanation combining the explanation of the first businessman image and the introduction video can be obtained; the method and the system enable the current user to more intuitively obtain the relevant information of the current house source concerned by the current user, and provide temperature service for the user.

According to the house source explanation method provided by the embodiment of the disclosure, the relevant information of the current house source is obtained according to the check request of the current user to the current house source; wherein, the related information comprises description text, at least one related broker and introduction video; a first lecture video generated based on the lecture text and relevant information of a first broker of the at least one relevant broker; processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video; obtaining a house source video explaining the current house source for the current user based on the second lecture video and the introduction video; the first person is confirmed through the house source related information, the second lecture video which is described by the first person image is obtained, the participation degree of brokers is enhanced, the service perception and the service experience of customers are improved, the user portrait is combined, the explanation of the house source is more fit with the requirements of the current users, and the efficiency of obtaining user information is improved.

As shown in fig. 2, based on the embodiment shown in fig. 1, step 104 may include the following steps:

step 1041, determining a first broker according to the contribution of the at least one relevant broker to the current house source.

Optionally, in this embodiment, one broker may be randomly extracted from at least one relevant broker corresponding to the current house source as the first broker, or one relevant broker with a higher contribution value is obtained as the first broker by ranking the contribution values of the at least one relevant broker to the current house source.

Step 1042, generating an introduction audio corresponding to the timbre of the first person according to at least one section of preset audio and the description text corresponding to the first person.

In this embodiment, at least one preset audio of the first broker lecture setting content is obtained in advance, and the content of the setting audio is not required to be consistent with the description text, and may be any content, and may specifically be set according to an application scenario; optionally, before the method of the present embodiment is applied, at least one preset audio segment in which at least one broker (for example, each broker of the at least one broker) speaks the same setting content may be stored, and the timbre of the first broker is obtained by performing timbre training on the at least one preset audio segment; by using the obtained timbre and the description text of the first epoch person through a TTS (TextToSpeech) speech synthesis technique, the introductory audio that describes the description text in the timbre of the first epoch person can be obtained.

And 1043, obtaining a first lecture video corresponding to the introduction audio based on the introduction audio and at least one preset action video corresponding to the first businessman.

The first lecture video comprises at least one description part, and the description part (for example, each description part) comprises at least one lecture data.

In this embodiment, in order to obtain a first lecture video including not only a first broker timbre but also a first broker image, a video of at least one broker (including the first broker) making at least one preset action is pre-captured, wherein the at least one action may include, but is not limited to: silence, call, think, point up, point down, etc.; the upper body (including at least the head and the upper limbs) of the broker is typically included in the preset action video; presetting the video length of the motion video to ensure the integrity of the motion (including the motion from beginning to end) to ensure the consistency of the connection among a plurality of preset motion videos, for example, setting the length of each preset motion video to be 6 seconds and the like, optionally, obtaining the preset motion video without limiting the mouth motion (usually, closing the mouth) of the broker; in the embodiment, at least one preset action video action is driven by the audio content in the introduction audio to obtain a first lecture video which is close to a real first broker lecture introduction text, wherein the first lecture video comprises at least one preset action, so that the first lecture video is more vivid and close to a real person image.

As shown in fig. 3, based on the embodiment shown in fig. 2, step 1043 may include the following steps:

step 301, processing at least one preset action video with the first duration based on the second duration of the introduction audio to obtain a processed video with the second duration.

Optionally, in order to replace the mouth in at least one preset motion video based on the introduction audio, the second duration needs to correspond to the first duration, and since it is desired to obtain the first lecture video corresponding to the introduction audio last, that is, the duration of the first lecture video is the second duration, the present embodiment processes at least one preset motion video to obtain the processed video.

Step 302, replacing the mouth movements in the processed video according to the introduction audio to obtain a first lecture video of the first person lecture introduction audio.

In the embodiment, only the mouth part in the processed video is replaced, and the rest part of the face in the video may have a change in resolution, but the overall expression, such as whether to blink or not, does not change. Therefore, the eye-blinking effect, the slight shaking of the head, the micro-expression effect of the facial muscles, the limb movement and the like are not changed, and the effect of falsifying and falsifying can be achieved. In the embodiment, the video of the first commented description audio is obtained only by replacing the mouth action, so that the obtained first lectured video is more real, the expression and the action of a real person during speaking can be more truly simulated, and the impression of a user is greatly improved.

Optionally, on the basis of the above embodiment, step 301 may include:

since the introduction audio of the second duration in the embodiment of the present disclosure introduces the current house source, the second duration is necessarily much longer than the first duration, and in this embodiment, at least one preset action video is combined according to the content in the introduction audio to obtain a processed video (which may include multiple identical preset action videos, for example, a preset action video including multiple silent actions).

The combining the preset action videos according to the content of the introduction audio may include, for example, the introduction audio includes a content of calling (e.g., hello, etc.), and at this time, the preset action video corresponding to the corresponding time period is the action video of calling; for consecutive preset motion videos with multiple segments of the same motion, at least one segment of the same preset motion video may be repeatedly played, and the repeated manner may include various manners, for example, including but not limited to: playing repeatedly from the first second to the last second of the preset action video; or playing the preset motion video from the first second to any time point, and then playing the preset motion video from the time point forward for multiple times in a circulating way until the corresponding time length is obtained; or, the preset action video is played from the first second to the last second in a forward sequence, and then is played from the last second to the first second in a reverse mode, and then is played circularly and continuously; or, the forward playing and the reverse playing are combined randomly; the back seeding in the circulation process can solve the difference of the mouth shapes of the connecting time points of the two circulations.

As shown in fig. 4, based on the embodiment shown in fig. 3, step 302 may include the following steps:

step 3021, obtaining at least one continuous frame of a first face image based on the processed video.

Since the duration of processing the video is the second duration, and therefore, the video includes at least one frame of video image, in this embodiment, the processing video is firstly decomposed into at least one frame of video image, and a frame of first face image (the first face image of the face part of the first person can be obtained by performing face recognition on the video image) is obtained based on the video image (including at least the face part image and the upper limb part image of the first person), respectively, and the face image included in the first face image is the face image of the first person.

Optionally, the face detection is performed on at least one frame of video image in the processed video to obtain at least one frame of first face image. The face detection can be based on a face detection network or other face detection methods in the prior art, and the purpose of the face detection in this embodiment is to remove the influence of the background in the video image on the subsequent operation, and only use the face part in the video image as the object of mouth replacement, so that the accuracy of mouth replacement is improved; the implementation does not limit the specific method of face detection, and only needs to realize the detection to obtain the face image.

And step 3022, replacing the mouth part in the at least one frame of first facial image based on the introduction audio to obtain at least one frame of second facial image.

For example, the mouth in each of the at least one frame of first face images is replaced based on the introductory audio.

In this embodiment, the mouth or the face including the part below the nose in the first face image is replaced based on the introduction audio, so that the mouth motion in the obtained second face image corresponds to the introduction audio, and optionally, the replacement process may be implemented in a frame-by-frame replacement manner.

And step 3023, connecting at least one frame of second human face image according to the sequence of at least one frame of first human face image to obtain a first lecture video.

Connecting at least one frame of second face image replacing the mouth according to the sequence of the first face image to obtain a first lecture video of which the mouth action corresponds to the complete introduction audio, wherein the time length of the first lecture video and the time length of the processed video are the same, and the face action in the first lecture video is completely the same as the processed video except the mouth, so that the real simulation on the micro expression and the action is realized, and therefore, the first lecture video is closer to the situation of the real target object (first person in the first era) in the introduction audio; when the voice content needs to be replaced, the mouth in the processed video only needs to be replaced correspondingly through the voice content needing to be replaced, and the effect of quickly generating the first lecture video is achieved.

Optionally, on the basis of the foregoing embodiment, step 3022 may include:

and segmenting the introduction audio to obtain at least one voice segment.

The voice segments correspond to the first face images, for example, each voice segment corresponds to each frame of the first face image in at least one frame of the first face image.

In this embodiment, the introduction audio may be divided into at least one voice segment with a certain time span according to the actual voice content, for example, each voice segment is 50ms, and in order to ensure the continuity of the mouth shape in the obtained first lecture video, an overlapping portion exists between two adjacent voice segments; alternatively, the received voice data may be a voice time domain signal or directly a voice feature, and when the received voice data is a voice time domain signal (waveform), the MFCC feature is extracted from the voice data (fourier transform of the waveform of a time window, and conversion of the time domain signal into a frequency domain signal), or the voice feature is extracted by a neural network, so as to replace a mouth in the first face image with the voice feature; the length of the corresponding speech segmentation may be determined according to a window of fourier transform.

And generating a second face image based on the first face image of the voice segment (for example, each voice segment) corresponding to the voice segment, and obtaining at least one frame of second face image.

And the face mouth shape in the second face image corresponds to the voice segment.

In this embodiment, the mouth of the corresponding frame of the first facial image is replaced by the speech segment, optionally, the mouth of the corresponding frame of the first facial image may be replaced based on a deep learning algorithm, for example, the speech segment and the first facial image corresponding to the speech segment are processed based on a neural network, so as to obtain the second facial image.

In an alternative example, a speech segment and a frame of a first face image are received through a neural network, a second face image replacing a mouth image is output, the second face image is different from the first face image only in that the speech segment of the mouth image corresponds to the speech segment of the mouth image, and before applying the neural network, optionally, the method further comprises:

and training the confrontation generation network by utilizing the sample data set.

The confrontation generating network comprises a neural network and a discrimination network, the sample data set comprises a plurality of pairs of sample data pairs, and each sample data pair comprises a sample voice segment and a sample face image of which the mouth corresponds to the sample voice segment.

Optionally, the neural network in this embodiment may be a generating network.

In the training of the embodiment, the sample face image is input into both the neural network and the discrimination network, whether the input image is generated by the neural network or a real image (the sample face image is used in the training) can be judged by the discrimination network in the confrontation generation network, the neural network and the discrimination network in the confrontation generation network are alternately trained, and then the trained neural network is used as a target network to generate the face image of which the mouth shape is close to the real condition and which corresponds to the voice segment.

Optionally, the training process may include:

inputting the sample data into a neural network in the confrontation generation network to obtain a generated image;

inputting the generated image and the sample face image in the sample data pair into a discrimination network to obtain a discrimination result;

and alternately training the discrimination network and the neural network based on the discrimination result to obtain the trained neural network.

The training process against the generated network is similar to the prior art, and the difference is only that the applied sample data is different, and the sample data pair in the embodiment includes: the training target is to enable the trained neural network to generate an image based on the input sample voice segment and the small difference between the output of the sample face image and the sample face image, and to judge whether the training target of the network is an image capable of accurately identifying whether the input image is real or not; the implementation of replacing the mouth image in the face image based on the neural network in the confrontation generating network provided by this embodiment is only an example, and only the mouth part in the first image needs to be replaced with the mouth corresponding to the voice segment in the implementation process.

As shown in fig. 5, based on the embodiment shown in fig. 1, step 106 may include the following steps:

step 1061, obtain the user portrait of the current user.

The user representation may include, but is not limited to, basic information of the current user, real estate preferences, and behavioral data, among others.

In this embodiment, the initial user representation data refers to a user representation obtained from history or external data when the user has not made an interactive question. Although the answers acquired in the interactive questions belong to the user portraits, the interactive questions need active operation of the user, the acquisition cost is high, and the experience is not influenced by too many questions, so that the initial user portrait data is also an important part in the user portrait; the user representation in this embodiment may be an initial user representation or a user representation supplemented with interactive questions after the current user has viewed some of the sources.

Step 1062, determining whether the user portrait affects the speaking point sequencing, if so, executing step 1063; otherwise, step 1064 is performed.

Optionally, the house purchase preferences may include, but are not limited to: one or more of price preference, total price lowest, total price highest, business district preference, living room preference, area preference, subway room preference, school district room preference, periphery preference, district preference, floor preference (not elevator), finishing requirements, elevator preference, floor age preference, house type preference (label), orientation preference, finishing-style preference, finishing-reinstallation intent, finishing-design preference, trip mode preference, and the like. And the information which is more concerned by the user in the house source is reflected.

The basic information in the user representation may include, but is not limited to: one or more items of information such as age, gender, marital status, mobile phone number, occupation, income condition, family people, whether children exist, child age, whether old people exist, whether cars exist, whether first cover exists, whether loan is available, house purchasing purpose, first payment budget, house purchasing qualification, loan mode and the like.

Behavioral data in a user representation may include, but is not limited to: one or more items of information such as number of listening commercial houses, number of listening cells, historical information of listening cells, number of listening houses, historical information of listening houses, number of viewing details pages of houses, number of viewing cells, number of viewing VR houses by oneself, houses viewed by a broker VR belt, houses viewed by an offline belt, cell familiarity and the like.

Step 1063, executing deletion and/or talk point data reordering processing on the first lecture video according to the user image; and obtaining the processed second lecture video.

Step 1064, directly using the original first lecture video as the second lecture video.

In this embodiment, when the user portrait of the current user shows that the current user does not care about some information, the corresponding content may be deleted in the first lecture video, for example: deleting the description part around the cell according to the value of the portrait label, and not explaining the description part which is not selected by the user; another example is: and (3) deleting the speaking points in the school education description part according to the values of the portrait _ house (type of school concerned) portrait label in the user portrait: school types not selected by the user do not appear in the lectures of the school education description section. The interest of the current user in the video is improved, and the phenomenon that the user loses the interest of continuously viewing due to the fact that the content which is not interested by the user is played is avoided; moreover, when the sorting of the talkback data is influenced according to the value of the portrait label, the talkback data is reordered, so that a user can preferentially hear more concerned contents, and the viewing interest of the video is improved; when the value of the portrait label does not affect the sorting of the speaking point data, the embodiment directly takes the first lecture video as the second lecture video to be played by the current user.

For example, in the following table 1, in this example, the portrait label of the user portrait is whether there is a car, the values of the portrait label include car and no car, the sequence of 3 speaking points (subway, high-speed entrance, bus station) included in the traffic information description section around the different-value cell is completely different, the table 1 uses the sequence number to represent the sequence of at least one speaking point when the value of the corresponding different portrait label is taken, taking the embodiment shown in table 1 as an example, when the value of the portrait label of the current user shows that there is a car, the sequence of the speaking points in the description section is: high-speed entrances, subways, bus stations; and when no vehicle exists, the talking sequence is as follows: subway, bus station, high-speed entrance.

Talking point	Wheeled vehicle	No vehicle
			Subway	2	1
High speed inlet	1	3
			Bus station	3	2

TABLE 1

For another example, in table 2 below, in this example, the portrait label of the user portrait is whether there is a child, values of the portrait label include children and no children, the sequence of 5 speaking points (school, traffic, shop, park, general _ hospital) included in the description part of the information around the different-value cell is completely different, the sequence of at least one speaking point in the value of the corresponding portrait label is represented by the sequence number, and the sequence of speaking points corresponding to the value of the corresponding portrait label can be obtained according to the sequence number.

TABLE 2

After the second lecture video is determined in the above embodiment, in the process of playing the house source video, the method further includes:

for at least one preset questioning node, determining a target question for the preset questioning node according to an extraction rule;

for example, for each preset questioning node in the at least one preset questioning node, a target question is determined for each preset questioning node according to an extraction rule.

And adjusting the extraction rule according to the feedback result of the current user so as to improve the listening time of the room source video.

When playing a house source video of a current house source for a current user, in order to obtain values of portrait tags included in more user portraits of the current user, information can be obtained in a question-answer interaction mode, optionally, at least one preset question node is set in the house source video (for example, a keyword is set, when the keyword is identified, a question is issued, different keywords can correspond to different questions), when one preset question node is identified in the process of playing the house source video, a question is obtained from at least one question corresponding to the preset question node according to a set extraction rule, the question is displayed in the video, the current user can select whether to answer the question, for example, whether the question needs to know the house type condition of the current house source or not, and according to whether the user answers or not, corresponding answers are provided for the current user, and playing contents in a subsequent house source video are adjusted (for example, delete content that is not of interest to the current user) to improve the time that the house source video is listened to by the current user and the conversion rate of the house source.

Optionally, determining a target question for a preset question node according to an extraction rule, including:

determining a ranking of portrait tags according to a weight value of at least one portrait tag included in a user portrait of a current user;

determining a target portrait label corresponding to a preset questioning node based on the sequence of the portrait labels;

the problem corresponding to the target image label is used as the target problem.

In this embodiment, the portrait tags (for example, including buying interest layer _ Intent (which may be old people + children, investment, old people, education, general purpose, self-residence, just need, improvement), child has _ child (which may be yes or no), child age child _ range, old people has _ old, whether the community is familiar with layer _ family _ resplock, periphery attention, school attention type school _ house, whether the community is subway _ house, number of rooms room _ cnt, whether there is elevator has _ elevator, decoration resolution _ style, frame tag frame _ label, floor type floor _ type, whether there is car has _ car, priority layer _ first, whether there is elevator _ door, whether there is display _ door, payment mode, and other situations may be determined according to the actual subway type, not listed here, values reasonably obtained by those skilled in the art are all values available in this embodiment) to set corresponding problems, so as to determine a value corresponding to a portrait label (for example, each portrait label), for example, set a problem on the intention of buying a house: what is your desire to buy (illustrated as an example only, is one of the at least one quiz template to which the portrait label corresponds)? The present embodiment determines the most concerned problem for the current user according to the weight ranking of at least one portrait label, and the weight value of at least one portrait label may be set according to the actual situation, for example, according to the following table 3, in which only the weight ranking results of ranking portrait labels according to weight are displayed, wherein the smaller the sequence number in the weight ranking, the larger the corresponding weight value.

buyer_feature	Weight sort order number
		buyer_Intent	1
has_child	4
		child_range	7
has_old	5
		buyer_familiar_resblock	3
surround	2
		school_house	8
subway_house	9
		room_cnt	11
has_elevator	10
		decoration_style	12
frame_label	13
		floor_type	14
has_car	6
		buyer_first	15
buyer_mortgate	16
		buyer_payment	17

TABLE 3

Optionally, before determining the ranking of portrait tags according to a weight value of at least one portrait tag included in a user portrait of a current user, further comprising:

and classifying at least one question corresponding to the current house source according to the corresponding portrait label.

The portrait tags correspond to questions, for example, each portrait tag corresponds to a question.

In this embodiment, the problem is to perfect the user portrait, so a problem is preset for the portrait label, and in this embodiment, the problem is first associated with the corresponding portrait label.

And performing first screening on at least one question based on the portrait label corresponding to the question determined by classification to obtain a first candidate question set.

Wherein the first set of candidate questions is empty or comprises at least one candidate question.

The screening of the first set of candidate questions may include: determining whether the portrait label corresponding to the problem has corresponding data in the user portrait of the current user according to classification;

removing the question from the at least one question when the corresponding data exists;

when there is no corresponding data, adding the question as a candidate question to the first set of candidate questions.

In this embodiment, it is unknown whether the user portrait of the current user is complete, when the user portrait is complete, at least one portrait label includes corresponding data (i.e., a value), at this time, the first candidate question set obtained through the first screening is empty, and at this time, a question does not need to be executed for the current user; but in general, the user image is incomplete, including at least one image tag without corresponding data, and the image tag without corresponding data is the first candidate problem set of the current user.

And carrying out second screening on the first candidate problem set according to the portrait label and the related information of the current house source to obtain a second candidate problem set.

Wherein the second set of candidate questions is empty or comprises at least one candidate question.

In this embodiment, the second filtering is implemented based on a constraint condition, where the constraint condition may include but is not limited to: determining that no question is asked or the like by combining the actual situation of the current house source or the current user, and deleting the candidate questions meeting the constraint condition, for example, no school exists around the current house source, and the questions corresponding to the portrait label concerning the school type can not be included in the second candidate question set; the second candidate question set obtained through the second screening may also be empty, for example, when the first candidate question set is empty, the second candidate question set is inevitably empty, or the first candidate question set is not empty, but through the second screening, it is found that the questions included in the first candidate question set may not be asked, and at this time, the second candidate question set is also empty; only when the second candidate question set is not empty, the question is issued to the current user, so that the user aversion caused by repeatedly asking the known questions to the user is avoided, and the probability of answering the questions by the user is improved.

In this embodiment, determining a ranking of portrait tags based on a weight value of at least one portrait tag included in a user portrait of a current user includes:

determining the ranking of at least one portrait label corresponding to the second candidate question set based on the weight values of the portrait labels corresponding to the candidate questions included in the second candidate question set.

In the embodiment, the number of portrait tags worth determining the weight values is reduced, and through twice screening, only the portrait tags corresponding to the portrait tags are sorted based on the portrait tags corresponding to the candidate questions in the second candidate question set, so that the processing efficiency is greatly improved, the portrait tags are more targeted, the probability of answering questions by a user is improved, and the probability of perfecting the portrait of the user is improved.

In an optional example, further comprising: and determining a question template from at least one question template corresponding to the determined target question.

Optionally, each question in the at least one question corresponds to at least one question template;

in order to avoid providing discordant questions for at least one different user, which results in mechanicalness and user's repugnance, the embodiment provides that at least one question template is set for the questions (for example, each question), and the same question is asked in different ways, so that the simulation of the question is enhanced, the user experience is closer to the real-person interaction, the user impression is improved, and the probability of the user for answering the question is improved; when a question template is determined for the template questions, a question template can be randomly selected from at least one question template for question asking, or the probability of the question being lifted is recorded after the question template is selected, and the question template is selected according to the corresponding probability.

Fig. 6 is a schematic flow chart of obtaining a target question according to an extraction rule in an example of the room source explanation method proposed by the present disclosure. As shown in fig. 6, starting from the identification of the preset question node, step 601, determining whether the number of questions included in the preset question node is greater than 0, and when the number of questions is not greater than 0, not performing question asking at the preset question node; when the number of questions is greater than 0, go to step 602; step 602, extracting corresponding questions according to current preset question nodes, and classifying the questions according to portrait labels of user portrait; step 603, judging whether the portrait label corresponding to each question has a value, and if so, invalidating the question; otherwise, go to step 604; step 604, the question is taken as a candidate question and is added into a first candidate question set; step 605, traversing all the problems in each portrait label and the house source related information; step 606, determining whether the candidate question in the first candidate question set does not satisfy the constraint condition, if so, taking the candidate question as a candidate question in the second candidate question set, and executing step 607, otherwise, the question is not asked; step 607, determining whether the number of candidate questions in the second candidate question set is greater than 0, if yes, the second candidate question set is valid, executing step 608, otherwise, the second candidate question set is invalidated; step 608, sorting the candidate questions according to the weight values of the portrait labels corresponding to the candidate questions; step 609, determining a plurality of question templates corresponding to the questions with the highest weight values corresponding to the portrait labels from the current second candidate question set; step 610, randomly determining a question template from a plurality of question templates for question asking.

In some optional embodiments, the adjusting the extraction rule according to the feedback result of the current user includes:

storing the target question and the feedback result, and determining the answer rate of the question based on the stored feedback result of at least one question;

The embodiment can further improve the portrait of the user based on the feedback result of the current user, and store the target question, so as to count the probability that the question of the question is answered, and adjust the weight value of the portrait label corresponding to the question with the probability that the question is answered, for example, the weight value of the portrait label corresponding to the question with low answer rate is reduced, and the probability that the next question is proposed is greatly reduced, so that the probability that the user answers the question is improved, and the impression of the user and the efficiency of the perfect portrait of the user are improved.

Optionally, the feedback results include answers or no answers;

before the target question and the feedback result are stored and the answer rate of the question is determined based on the stored feedback result of at least one question, the method further comprises the following steps:

determining whether the current user answers the target question;

responding to the current user to answer the target question, providing a corresponding answer for the current user, supplementing a user portrait, and updating a weight value of corresponding talk point data according to the supplemented user portrait; broadcasting the talk point data according to the updated weight value;

and responding to the target question which is not answered by the current user, waiting for a set time or providing a corresponding answer for the current user, and recording that the current user does not answer.

Fig. 7 is a schematic flowchart illustrating a question interaction with a current user in an example of a room source explanation method proposed by the present disclosure. As shown in FIG. 7, step 701, obtain an initial user representation of a current user; step 702, judging whether the initial user portrait affects the speaking point sequence in the house source video, if so, executing step 703 after adjusting the speaking point sequence, otherwise, executing step 703 after keeping the initial speaking point sequence; step 703, playing a room source video; step 704, in response to identifying the preset question node, extracting the question according to the rule (optionally extracting the question according to the flow shown in fig. 6); 705, judging whether the user answers or not, if so, replying the client, perfecting the user portrait, adjusting the corresponding weight of the talk point according to the answer, sequencing the talk point according to the adjusted weight, playing the house source video, and executing 706; otherwise, waiting for a set time and/or replying to the client, recording that the client does not answer, adjusting a question policy (for example, reducing a weight value of the portrait label corresponding to the question, and the like), playing the house source video according to a default sequence, and executing step 706; step 706, storing the extracted questions and/or answer rates; step 707, adjusting a problem extraction strategy; step 708, storing at least one of the adjusted speaking point weight, the withdrawal rate and the listening duration; and step 709, adjusting the setting of the speaking point weight, and ending. The client in this embodiment is equivalent to the user.

In some optional embodiments, the information related to the current house source further comprises: at least one bright spot image and bright spot description information corresponding to the bright spot image;

the method provided by the embodiment further comprises the following steps:

determining a bright point explanation video corresponding to at least one bright point image based on bright point description information corresponding to the at least one bright point image and at least one preset action video corresponding to a first businessman;

sequencing at least one highlight display video based on a user image of a current user;

and displaying at least one highlight display video according to the sequence.

Some house sources have outstanding characteristics, such as north-south permeability, full-clear pattern, traffic convenience and the like; corresponding images and description information are corresponding to the characteristics, in order to highlight the main characteristics of the house source and attract the user to continuously listen to the house source video, in the embodiment, the highlight information of the current house source is displayed before the house source video is played, the display process not only includes the highlight images and the highlight description information (text information), but also includes the highlight explanation video corresponding to the first businessman, the embodiment sets the highlight image display set duration (for example, each image is displayed for 5 seconds and the like), at least one preset action video corresponding to the first businessman is driven through the highlight description information in the display duration (the process is similar to the process of determining the introduction audio based on the introduction text and driving the preset action video based on the introduction audio, and the preset action video is not repeated herein), the first broker description highlight description information corresponding to the set duration can be obtained and is matched with the introduction action (for example, directional, etc.), the bright spot explanation video that will only include the broker obtains bright spot demonstration video in the bright spot explanation video imbeds the show image of bright spot image and bright spot description information, only the broker makes the action in this video, bright spot image and bright spot description information are unchangeable, in addition, can also include the point of speaking that bright spot description information corresponds, if correspond a plurality of points of speaking, choose one to speak the point and show can, this embodiment only shows the bright spot demonstration video of setting for quantity in order to improve user interest degree, when the quantity of bright spot demonstration video is greater than this setting for quantity, the random selection wherein set for quantity bright spot demonstration video show can show.

In some optional embodiments, before obtaining the relevant information of the current house source according to the viewing request of the current user for the current house source, the method further includes:

determining the authenticity of the related data by matching the related data input by at least one terminal corresponding to at least one house source;

In the embodiment, in order to perfect basic data, a data acquisition mechanism of platform big data extraction and broker supplementation is established, wherein the platform big data extraction mainly performs acquisition source of required knowledge points and arrangement of calculation rules; however, the platform has partial data missing, so that a tool for improving data by a broker is added, a terminal (for example, a mobile terminal) interface is improved for the broker, the data input by at least one broker through at least one terminal to a house source is accepted, certainly, the data input by the broker cannot be directly used as related information of the house source, and the authenticity of the data needs to be determined, the authenticity determination can be based on the set number (the number can be set according to the actual situation, for example, 5 data and the like) of the same talk point input by the broker through the terminal, whether the data are the same or different in the set range is determined, at this time, the input related data are determined to be real, the true experiment is realized, and the relevant data subjected to the true experiment are stored in the related information of the house source to improve the related information of the house source; the broker can improve the contribution value of the house source by inputting the relevant data of the house source subjected to the truth test, and further improve the probability of explaining the house source.

Optionally, the step 108 may include: and performing corresponding processing (such as deletion and/or sequence adjustment and the like) on the introduction video based on the second teaching video, and embedding the second teaching video into the processed introduction video to obtain the house source video.

Optionally, in the introduction video, a corresponding special effect animation can be set according to several specific advantages of the current house source, so as to improve the ornamental value of the house source video. For example, as shown in fig. 8a, a special effect animation as shown in the figure (the special effect animation is only shown in the form of an image limited by the display) is set for the features of the full-clear pattern of one house source; as shown in fig. 8b, the north-south permeability of one room source is characterized by the special effect animation as shown in the figure (the special effect animation is only shown in the form of an image by being limited to be displayed); as shown in fig. 8c, the characteristic of the dynamic and static separation of one room source is set as a special effect animation as shown in the figure (in the figure, dynamic and static areas can be distinguished through different colors, and the special effect animation is only shown in an image form limited by display); as shown in fig. 8d1-2, the special effect animation is shown for the characteristics of the Ming kitchen and Ming toilet of a house source (the special effect animation is only shown in the form of images by being limited by the display); the embodiment is limited to show only a few special effect animations by space, but the showing is not used for displaying the disclosure, and it also belongs to the protection scope of the disclosure to set corresponding special effect animations for other characteristics that are more prominent to house resources.

Any of the methods of house source explanation provided by the embodiments of the present disclosure may be performed by any suitable device having data processing capabilities, including but not limited to: terminal equipment, a server and the like. Alternatively, any of the house source explanation methods provided by the embodiments of the present disclosure may be executed by a processor, for example, the processor may execute any of the house source explanation methods mentioned by the embodiments of the present disclosure by calling a corresponding instruction stored in a memory. And will not be described in detail below.

Exemplary devices

Fig. 9 is a schematic structural diagram of an atrioventricular source explanation device according to an exemplary embodiment of the present disclosure. As shown in fig. 9, an apparatus provided in an embodiment of the present disclosure includes:

the house source information obtaining module 91 is configured to obtain relevant information of a current house source according to a request of a current user for viewing the current house source.

A first lecture video module 92 for generating a first lecture video based on the lecture text and information about a first broker of the at least one relevant broker.

The second lecture video module 93 is configured to process the first lecture video based on a user image corresponding to the current user, and obtain a processed second lecture video.

And the house source explanation module 94 is configured to obtain a house source video explaining the current house source for the current user based on the second explanation video and the introduction video.

According to the house source explanation device provided by the embodiment of the present disclosure, relevant information of a current house source is obtained according to a current user's request for checking the current house source; wherein, the related information comprises description text, at least one related broker and introduction video; a first lecture video generated based on the lecture text and relevant information of a first broker of the at least one relevant broker; processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video; obtaining a house source video explaining the current house source for the current user based on the second lecture video and the introduction video; the first person is confirmed through the house source related information, the second lecture video which is described by the first person image is obtained, the participation degree of brokers is enhanced, the service perception and the service experience of customers are improved, the user portrait is combined, the explanation of the house source is more fit with the requirements of the current users, and the efficiency of obtaining user information is improved.

Optionally, a first lecture video module 92, specifically configured to determine a first broker according to a contribution value of at least one relevant broker to the current house source; generating introduction audio corresponding to the tone of the first menstrual person according to at least one section of preset audio and the description text corresponding to the first menstrual person; obtaining a first lecture video corresponding to the introduction audio based on the introduction audio and at least one preset action video corresponding to the first person; the first lecture video comprises at least one description part, and the description part comprises at least one lecture data.

Optionally, when the first lecture video corresponding to the introduction audio is obtained based on the introduction audio and the at least one preset action video corresponding to the first parent person, the first lecture video module 92 is configured to process the at least one preset action video with the first duration based on the second duration of the introduction audio, so as to obtain a processed video with the second duration; mouth movements in the processed video are replaced according to the introduction audio, and a first lecture video of the first person lecture introduction audio is obtained.

Optionally, the second lecture video module 93 is specifically configured to obtain a user representation of the current user; determining whether a user representation affects ranking of the speaking point data; in response to the user portrait affecting the sorting of the talk point data, deleting and/or reordering the talk point data of the first lecture video according to the user portrait; and obtaining the processed second lecture video.

Optionally, the apparatus provided in this embodiment further includes:

the node identification module is used for determining at least one preset questioning node included in the house source video based on the house source video and the description text;

the question extraction module is used for determining a target question for at least one preset question node according to an extraction rule;

Optionally, the question extraction module is specifically configured to determine a ranking of portrait tags according to a weight value of at least one portrait tag included in a user portrait of a current user; determining a target portrait label corresponding to a preset questioning node based on the sequence of the portrait labels; the problem corresponding to the target image label is used as the target problem.

Optionally, the question extraction module is further configured to classify at least one question corresponding to the current house source according to the corresponding portrait label; wherein, the portrait label has corresponding relation with the question; performing first screening on at least one question based on the portrait label corresponding to the question determined by classification to obtain a first candidate question set; wherein the first set of candidate questions is empty or comprises at least one candidate question; performing second screening on the first candidate problem set according to the portrait label and the related information of the current house source to obtain a second candidate problem set; wherein the second set of candidate questions is empty or comprises at least one candidate question;

the question extraction module is used for determining the ranking of at least one portrait label corresponding to the second candidate question set based on the weight value of the portrait label corresponding to the candidate question included in the second candidate question set when determining the ranking of the portrait labels according to the weight value of at least one portrait label included in the user portrait of the current user.

Optionally, the question extraction module is configured to determine, when at least one question is first screened based on the portrait label corresponding to the question determined by classification to obtain a first candidate question set, whether there is corresponding data in the user portrait of the current user according to the portrait label corresponding to the question determined by classification; removing the question from the at least one question when the corresponding data exists; when there is no corresponding data, adding the question as a candidate question to the first set of candidate questions.

Optionally, the question extracting module is further configured to determine a question template from at least one question template corresponding to the determined target question.

Optionally, the feedback results include answers or no answers;

the rule adjusting module is also used for determining whether the current user answers the target question or not; responding to the current user to answer the target question, providing a corresponding answer for the current user, supplementing a user portrait, and updating a weight value of corresponding talk point data according to the supplemented user portrait; playing the talk point data according to the updated weight value; and responding to the target question which is not answered by the current user, waiting for a set time or providing a corresponding answer for the current user, and recording that the current user does not answer.

the apparatus provided in this embodiment further includes:

the bright point video module is used for determining a bright point explanation video corresponding to the bright point image based on the bright point description information corresponding to the at least one bright point image and the at least one preset action video corresponding to the first businessman; determining a bright point display video based on the bright point explanation video, the bright point image and the bright point description information; sequencing at least one highlight display video based on a user image of a current user; and displaying at least one highlight display video according to the sequence.

Optionally, the apparatus provided in this embodiment further includes:

the information acquisition module is used for receiving at least one piece of relevant data corresponding to at least one house source in a database input by at least one terminal; determining the authenticity of the related data by matching the related data input by at least one terminal corresponding to at least one house source; and storing the relevant data with the authenticity determined as real into the relevant information of the corresponding house source in the database.

Exemplary electronic device

Next, an electronic apparatus according to an embodiment of the present disclosure is described with reference to fig. 10. The electronic device may be either or both of the first device 100 and the second device 200, or a stand-alone device separate from them that may communicate with the first device and the second device to receive the collected input signals therefrom.

FIG. 10 illustrates a block diagram of an electronic device in accordance with an embodiment of the disclosure.

As shown in fig. 10, the electronic device 10 includes one or more processors 11 and memory 12.

The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

Memory 12 may include one or more computer program products that may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory. The volatile memory may include, for example, Random Access Memory (RAM), cache memory (cache), and/or the like. The non-volatile memory may include, for example, Read Only Memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer readable storage medium and executed by the processor 11 to implement the above-described methods of origin explanation of the various embodiments of the present disclosure and/or other desired functions. Various contents such as an input signal, a signal component, a noise component, etc. may also be stored in the computer-readable storage medium.

In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other form of connection mechanism (not shown).

For example, when the electronic device is the first device 100 or the second device 200, the input device 13 may be a microphone or a microphone array as described above for capturing an input signal of a sound source. When the electronic device is a stand-alone device, the input means 13 may be a communication network connector for receiving the acquired input signals from the first device 100 and the second device 200.

The input device 13 may also include, for example, a keyboard, a mouse, and the like.

The output device 14 may output various information including the determined distance information, direction information, and the like to the outside. The output devices 14 may include, for example, a display, speakers, a printer, and a communication network and its connected remote output devices, among others.

Of course, for simplicity, only some of the components of the electronic device 10 relevant to the present disclosure are shown in fig. 10, omitting components such as buses, input/output interfaces, and the like. In addition, the electronic device 10 may include any other suitable components depending on the particular application.

Exemplary computer program product and computer-readable storage Medium

In addition to the above-described methods and apparatus, embodiments of the present disclosure may also be a computer program product comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the method of atria narrative according to various embodiments of the present disclosure described in the "exemplary methods" section of this specification above.

The computer program product may write program code for carrying out operations for embodiments of the present disclosure in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server.

Furthermore, embodiments of the present disclosure may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform the steps in the method of house source explanation according to various embodiments of the present disclosure described in the "exemplary methods" section above in this specification.

The computer-readable storage medium may take any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The foregoing describes the general principles of the present disclosure in conjunction with specific embodiments, however, it is noted that the advantages, effects, etc. mentioned in the present disclosure are merely examples and are not limiting, and they should not be considered essential to the various embodiments of the present disclosure. Furthermore, the foregoing disclosure of specific details is for the purpose of illustration and description and is not intended to be limiting, since the disclosure is not intended to be limited to the specific details so described.

In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts in the embodiments are referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The block diagrams of devices, apparatuses, systems referred to in this disclosure are only given as illustrative examples and are not intended to require or imply that the connections, arrangements, configurations, etc. must be made in the manner shown in the block diagrams. These devices, apparatuses, devices, systems may be connected, arranged, configured in any manner, as will be appreciated by those skilled in the art. Words such as "including," "comprising," "having," and the like are open-ended words that mean "including, but not limited to," and are used interchangeably therewith. The words "or" and "as used herein mean, and are used interchangeably with, the word" and/or, "unless the context clearly dictates otherwise. The word "such as" is used herein to mean, and is used interchangeably with, the phrase "such as but not limited to".

The methods and apparatus of the present disclosure may be implemented in a number of ways. For example, the methods and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order for the steps of the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless specifically stated otherwise. Further, in some embodiments, the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine-readable instructions for implementing the methods according to the present disclosure. Thus, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

It is also noted that in the devices, apparatuses, and methods of the present disclosure, each component or step can be decomposed and/or recombined. These decompositions and/or recombinations are to be considered equivalents of the present disclosure.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the disclosure to the form disclosed herein. While a number of example aspects and embodiments have been discussed above, those of skill in the art will recognize certain variations, modifications, alterations, additions and sub-combinations thereof.

Claims

1. A house source explanation method, comprising:

2. The method of claim 1, wherein the generating a first lecture video based on the lecture text and information about a first broker of the at least one relevant broker comprises:

3. The method of claim 2, wherein obtaining a first lecture video corresponding to the introduction audio based on at least one preset action video corresponding to the introduction audio and the first person comprises:

4. The method according to claim 2 or 3, wherein the processing the first lecture video based on the user image corresponding to the current user to obtain a processed second lecture video comprises:

obtaining a user representation of the current user;

5. The method of any of claims 1-4, further comprising:

6. The method according to claim 5, wherein the determining a target question for the preset questioning node according to an extraction rule comprises:

7. The method of claim 6, further comprising, prior to determining the ranking of the portrait tags based on a weight value of at least one portrait tag included in a user portrait of the current user:

8. An origin explanation device, characterized by comprising:

9. A computer-readable storage medium, characterized in that the storage medium stores a computer program for executing the house source explanation method according to any one of claims 1 to 7.

10. An electronic device, characterized in that the electronic device comprises:

a processor;

a memory for storing the processor-executable instructions;

the processor is configured to read the executable instructions from the memory and execute the instructions to implement the method of any of the above claims 1-7.