Live video processing method and device, live server and storage medium

Abstract

The invention provides a live video processing method and device, a live server and a storage medium, and the method comprises the steps: responding to a target virtual gift sent by an audience user in a live room, and monitoring first feedback information for live content in the live room after the audience user sends the target virtual gift; in response to the monitored forward feedback information in the first feedback information, determining the first live broadcast content corresponding to the forward feedback information in the first feedback information as target live broadcast content; and based on the gift display dynamic effect of the target virtual gift and the target live broadcast content, generating a recorded video and sending the recorded video to the audience user. According to the mode, the gift display dynamic effect of the target virtual gift in the live video and the fragments of the live content loved by the audience users can be automatically extracted and sent to the audience users, automatic editing of the gift display dynamic effect and the video fragments of the gift display dynamic effect expressing the appreciation process to the audience users by an anchor can be achieved, the accuracy and quality of video extraction are improved, and the user experience is improved. And the experience of audience users is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of live broadcast, and in particular to a live broadcast video processing method and device, a live broadcast server and a storage medium. BACKGROUND

[0002] In a live broadcast process, a viewer user can send a gift to a host, and the host can perform a thank-you performance in the form of language or a program to express thanks to the viewer user who sent the gift. In related technologies, a screen recording can be performed in the above process to record the viewer user sending a virtual gift and a segment of the performance content that the viewer user likes, but the accuracy and quality of the screen recording content are low, and the screen recording content can be incomplete or lengthy, which affects the experience of the viewer user. SUMMARY

[0003] Therefore, the present application aims to provide a live broadcast video processing method and device, a live broadcast server and a storage medium to automatically extract a video segment of a viewer user sending a gift and a live broadcast content that the viewer user likes in a live broadcast room, improve the accuracy and quality of video extraction, and improve the experience of the viewer user.

[0004] In a first aspect, an embodiment of the present application provides a live broadcast video processing method, which is applied to a live broadcast server. The method comprises the following steps: in response to a viewer user sending a target virtual gift in a live broadcast room, listening to first feedback information of live broadcast content in the live broadcast room after the viewer user sends the target virtual gift; in response to listening to positive feedback information in the first feedback information, determining first live broadcast content corresponding to the positive feedback information in the first feedback information as target live broadcast content; generating a recording video based on a gift display effect of the target virtual gift and the target live broadcast content; and sending the recording video to the viewer user.

[0005] In a second aspect, an embodiment of the present application provides a live broadcast video processing device, which is arranged in a live broadcast server. The device comprises the following modules: a first listening module, which is configured to listen to first feedback information of live broadcast content in a live broadcast room after a viewer user sends a target virtual gift in the live broadcast room in response to the viewer user sending the target virtual gift; a first determining module, which is configured to determine first live broadcast content corresponding to positive feedback information in the first feedback information as target live broadcast content in response to listening to the positive feedback information in the first feedback information; and a first sending module, which is configured to generate a recording video based on a gift display effect of the target virtual gift and the target live broadcast content, and send the recording video to the viewer user.

[0006] In a third aspect, an embodiment of the present application provides a live broadcast server, which comprises a processor and a memory. The memory stores machine executable instructions that can be executed by the processor. The processor executes the machine executable instructions to implement the live broadcast video processing method described above.

[0007] In a fourth aspect, an embodiment of the present application provides a machine readable storage medium, which stores machine executable instructions, and when the machine executable instructions are invoked and executed by a processor, the machine executable instructions cause the processor to implement the live video processing method.

[0008] The embodiments of the present application have the following beneficial effects:

[0009] The live video processing method, device, live server and storage medium, in response to the target virtual gift sent by the audience user in the live room, listen to the first feedback information of the live content in the live room after the target virtual gift is sent by the audience user; in response to the positive feedback information in the first feedback information, the first live content corresponding to the positive feedback information in the first feedback information is determined as the target live content; based on the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user. In this way, after the audience user sends the target virtual gift in the live room, the live server will listen to the feedback information of the live content in the live room after the target virtual gift is sent by the audience user, and after the positive feedback information is listened to, the target live content corresponding to the positive feedback information is obtained, and finally, the record video is generated according to the gift display effect of the target virtual gift and the target live content, and the record video is sent to the audience user. This way, by listening to the feedback information of the live content in the live room after the target virtual gift is sent by the audience user, the gift display effect of the target virtual gift and the video clip of the live content liked by the audience user in the live video are automatically extracted, and the record video is generated and sent to the audience user. The automatic editing of the video clip of the gift display effect and the process of the anchor user expressing thanks to the audience user can be realized, the accuracy and quality of video extraction are improved, and the experience of the audience user is improved.

[0010] Other features and advantages of the present application will be further described in the following description, and will be apparent to those skilled in the art from the following description and specific examples. The objects and other advantages of the present application will be realized and achieved by the structure particularly indicated in the description, claims and drawings.

[0011] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to make the technical solutions in the specific embodiments or the prior art of the present application clearer, the accompanying drawings needed in the specific embodiments or prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0013] Figure 1 A flow chart of a live video processing method provided by an embodiment of the present application;

[0014] Figure 2 A flow chart of a live video processing method provided by an embodiment of the present application;

[0015] Figure 3 A schematic diagram of a live video processing device provided by an embodiment of the present application;

[0016] Figure 4 A schematic diagram of a live server provided by an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the technical solutions in the specific embodiments or the prior art of the present application clearer, the accompanying drawings needed in the specific embodiments or prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0018] Based on this, the live video processing method, device, live server and storage medium provided by the embodiments of the present application can be applied to the live technology field.

[0019] In a possible implementation, the embodiments of the present application provide a live video processing method, which is applied to a live server, as shown in Figure 1 The live video processing method comprises the following steps:

[0020] Step S102, in response to the target virtual gift sent by the audience user in the live room, listening to the first feedback information of the live content in the live room after the target virtual gift is sent by the audience user;

[0021] It can be understood that the above audience user refers to the audience end in the live client, the anchor user refers to the anchor end in the live client, and during the live process, the audience user can send the corresponding virtual gift to the anchor user in the live room by consuming virtual resources such as points and gold coins. The above target virtual gift can be any virtual gift, or one or more virtual gifts that meet specified conditions, for example, the target virtual gift can be a virtual gift whose gift consumption resource exceeds a specified threshold, or a plurality of virtual gifts specified in advance. The above first feedback information is feedback information made by the audience user after sending the target virtual gift to the anchor user in the live room, and the first feedback information includes the text or language evaluation of the audience user to the anchor behavior, which can be sent through the barrage or a specified feedback interface. The first feedback information also includes the behavior of the audience user sending virtual gifts to the anchor user, and the behavior of sending virtual gifts is also a kind of feedback information, which expresses the recognition and love for the anchor behavior.

[0022] In one way, the audience user enters the live room by logging in the live account, and after determining the virtual gift to be sent, sends a gift sending instruction to the live server, the gift sending instruction containing the gift identifier of the virtual gift and the gift consumption resource corresponding to the virtual gift. After the live server receives the gift sending instruction, it identifies the gift consumption resource in the gift sending instruction, removes the virtual gift consumption resource from the total amount of virtual resources held by the audience user, and sends the virtual gift corresponding to the gift identifier to the anchor user. If the live server identifies that the gift consumption resource exceeds a specified threshold, it determines that the audience user has sent the target virtual gift in the live room, and the live server listens to the first feedback information of the live content in the live room after the audience user sends the target virtual gift.

[0023] Step S104, in response to listening to the positive feedback information in the first feedback information, the first live content corresponding to the positive feedback information in the first feedback information is determined as the target live content;

[0024] It can be understood that after the audience user sends the target virtual gift in the live room, the anchor user may make some specified behaviors in the live room to express thanks to the audience user, which can be a language thanksgiving behavior for the target virtual gift and a performance behavior for the target virtual gift, or only one of a language thanksgiving behavior for the target virtual gift or a performance behavior for the target virtual gift. The performance behavior for the target virtual gift includes singing performance for the target virtual gift and dance performance for the target virtual gift. Of course, the anchor user may not make specified behaviors for the target virtual gift, but continue to show some regular live content, such as explanation of goods, other program performance, etc.

[0025] In one mode, if the anchor user has a specified behavior in the live room for the target virtual gift, the live server can identify the specified behavior and provide a feedback interface for the specified behavior to the audience user, and the audience user can send feedback information for the specified behavior as the first feedback information through the feedback interface, and the live server obtains the evaluation of the specified behavior of the audience user through the feedback interface. For example, when it is identified that the anchor user sings and dances for the target virtual gift, the live server provides a feedback interface in the form of a pop-up window, and a scoring pop-up window is popped up in the live room content interface displayed on the terminal device of the audience user, and the singing and dancing behavior of the anchor user for the target virtual gift is displayed in the scoring pop-up window, and each behavior corresponds to a scoring interval, and the audience user can score each specified behavior of the anchor in the provided scoring interval according to his own will. Alternatively, during the singing and dancing performance of the anchor user for the target virtual gift, the live server controls the terminal device of the audience user to pop up an option pop-up window in the displayed live room content interface, and the audience user can select from multiple options such as “dissatisfied”, “average”, “satisfied”, and the like, to quickly evaluate the specified behavior of the anchor.

[0026] In another mode, the live server can also obtain the first feedback information for the anchor behavior through the barrage information. For the live content performed by the anchor, which belongs to the specified behavior or does not belong to the specified behavior, the live server can identify the feedback information of the audience user through the barrage system.

[0027] In another mode, the live server can also determine the first feedback information of the audience user for the live content through the gift sending instruction. It can be understood that the audience user can express recognition and love for the anchor behavior by sending virtual gifts. If the audience user sends a virtual gift to the anchor user during the anchor's specified behavior or other live content, the client of the audience user sends a gift sending instruction to the live server, the live server receives the gift sending instruction, determines that the audience user has sent a virtual gift to the anchor user, and the live server can determine that the audience user gives a positive feedback for the specified behavior or other live content that is occurring.

[0028] The above-mentioned positive feedback information is the positive feedback given by the audience user for the live content, indicating recognition and love for the live content. Here, the live content can be the specified behavior of the anchor for the target virtual gift, or the anchor behavior that does not belong to the specified behavior. The above-mentioned first live content is the live content for which the positive feedback information in the live room is directed after the audience user sends the target virtual gift. The above-mentioned target live content refers to the live content that is intended to be sent to the audience user and is loved by the audience. The target live content can be the first live content.

[0029] Herein, the positive feedback information in the first feedback information includes the specified barrage information or the feedback information including the specified feedback keywords obtained through the feedback interface; the positive feedback information in the first feedback information further includes the behavior of sending virtual gifts of the audience user.

[0030] Specifically, for the barrage information, some positive and praising words or phrases can be defined in advance as specified barrage keywords, such as the words “good” and “great”, and the audience user sends the target virtual gift, the live server listens to the barrage information initiated by the audience user, if the specified barrage keywords appear in the barrage information, it is determined that the barrage information is positive feedback information, at this time, the time when the audience user sends the barrage information is obtained, the behavior of the anchor in the live room corresponding to the time is determined, the live content of the anchor user when the anchor behavior occurs is determined as the first live content, and the first live content is determined as the target live content.

[0031] For the feedback information obtained through the feedback interface, the live server provides the feedback interface for the specified behavior of the anchor user to the audience user when the live server recognizes that the anchor user has the specified behavior for the target virtual gift. For example, for the feedback interface in the form of a scoring pop-up window, all score values in the scoring interval higher than a preset score threshold are specified as feedback keywords, and when the score input by the audience user is higher than the preset score threshold, it is determined that the feedback information obtained through the feedback interface is positive feedback information. For the feedback interface in the form of an option pop-up window, when the “satisfied” option selected by the audience user is received, the “satisfied” option selected by the audience user is specified as the specified feedback keyword, and it is determined that the feedback information obtained through the feedback interface is positive feedback information. After determining that the positive feedback information of the user for the specified behavior is listened to through the feedback interface, the corresponding specified behavior is obtained from the feedback interface, the live content of the anchor user when the specified behavior occurs is determined as the first live content, and the first live content is determined as the target live content.

[0032] The positive feedback information further includes the behavior of the audience user continuing to send gifts after sending the target virtual gift. It can be understood that if the audience user continues to send virtual gifts during the process of the anchor performing the specified behavior for the target virtual gift, whether the virtual gift consumes many or few gift resources, it is an approval and love for the anchor behavior, therefore, the live server can automatically confirm that the positive feedback information is received, determine the anchor behavior being performed by the audience user when sending the gift, determine the live content of the anchor user when the anchor behavior occurs as the first live content, and determine the first live content as the target live content.

[0033] In this step, after the live broadcast server listens to the positive feedback information in the first feedback information through various channels, the live broadcast server obtains the first live broadcast content corresponding to the positive feedback information, determines the first live broadcast content as target live broadcast content, and the target live broadcast content is the live broadcast content that the audience user likes.

[0034] In step S106, a recording video is generated based on the gift display effect of the target virtual gift and the target live broadcast content, and the recording video is sent to the audience user.

[0035] The live broadcast server can identify the anchor behavior occurring in the live broadcast room and mark the start and end times of various behaviors in the live broadcast room, and record the live broadcast content.

[0036] Specifically, the sending time point of the target virtual gift is marked.

[0037] The live broadcast server can identify the language thanksgiving behavior occurring for the target virtual gift, and record the start and end time points and the live broadcast content of the language thanksgiving behavior. In one way, the live broadcast server can identify the voice sent by the anchor user in real time, and when the account nickname of the audience user sending the target virtual gift, the gift name of the target virtual gift, or the keywords such as "thank you" and "thank you" are mentioned in the voice, it is determined that the anchor user has language thanksgiving behavior for the target virtual gift. The live broadcast server marks the time point when the sentence containing the keywords appears and the time point when the sentence ends, obtains the start and end time of the language thanksgiving behavior, records the live broadcast content corresponding to the start and end time of the language thanksgiving behavior as the live broadcast content of the anchor user's language thanksgiving behavior for the target virtual gift.

[0038] The live broadcast server can also identify the singing performance behavior of the anchor user, and record the start and end time points and the live broadcast content corresponding to the behavior. In one way, the voice of the anchor user in the live broadcast room can be identified according to the anchor tone, and the song can be identified in real time. When the song is identified, it is determined that the anchor user is singing. The live broadcast server marks the time point when the anchor user starts singing and the time point when the anchor user stops singing, obtains the start and end time of the singing performance, records the live broadcast content corresponding to the start and end time of the singing performance as the live broadcast content of the anchor user's singing performance behavior.

[0039] The live streaming server can also identify the dance performance behavior of the anchor user. In one approach, the live streaming platform can also use computer vision technology to analyze the video stream in the live streaming room. The live streaming server can detect the anchor's actions and gestures in the live streaming content by performing real-time analysis on the video stream. The live streaming server can use a gesture estimation algorithm to determine whether the anchor is dancing. After determining that the anchor user is dancing, the live streaming server can mark the time point at which the anchor user starts dancing and the time point at which the anchor user ends the dance. The live streaming server can obtain the start and end times of the dance performance. The live streaming server can record the live streaming content corresponding to the start and end times of the dance performance as the live streaming content in which the anchor user performs the dance.

[0040] The gift display effect of the target virtual gift described above is a gift special effect of the target virtual gift displayed on the interface of the audience user, which is obtained by the live streaming server from the live streaming video.

[0041] Finally, the video clip of the target virtual gift display effect and the video clip of the target live streaming content are cut from the live streaming video, and a recording video is generated by recombination. After the end of the current live streaming session or after the anchor user touches the video sending control, the recording video is automatically sent to the audience user, thereby realizing personalized video clipping for the audience user.

[0042] Specifically, in one approach, if the target live streaming content includes the language thanksgiving behavior of the anchor user for the target virtual gift and the performance behavior for the target virtual gift, a first video clip containing the gift display effect of the target virtual gift is obtained. According to the start and end times of the language thanksgiving behavior of the target virtual gift and the start and end times of the performance behavior for the target virtual gift, the live streaming content in which the language thanksgiving behavior occurs is extracted as a second video clip, and the live streaming content in which the performance behavior for the target virtual gift occurs is extracted as a third video clip. The recording video is generated in the order of the first video clip, the second video clip, and the third video clip, and the recording video is sent to the audience user.

[0043] Correspondingly, if the target live content contains the language thanksgiving behavior of the anchor user for the target virtual gift or the performance behavior for the target virtual gift, a first video clip containing the gift display effect of the target virtual gift is acquired, and a second video clip of the live content in which the language thanksgiving behavior occurs or the live content in which the performance behavior for the target virtual gift occurs is extracted according to the start and end time of the language thanksgiving behavior or the start and end time of the performance behavior for the target virtual gift, and a record video is generated according to the first video clip and the second video clip, and the record video is sent to the audience user.

[0044] In another way, if the anchor user in the live room does not have the specified behavior for the target virtual gift, in this case, the target live content does not contain the specified behavior for the target virtual gift, and at this time, the target live content can be a regular performance behavior such as singing, dancing, playing, etc. that the audience user likes. At this time, a second video clip containing the regular performance behavior is generated according to the start and end time of the regular performance behavior in the target live content, and a first video clip containing the gift display effect of the target virtual gift is generated, and a record video is generated according to the arrangement order of the first video clip and the second video clip.

[0045] In another way, in the current live session in the live room, if the audience user sends multiple target virtual gifts, the gift consumption resources of all the target virtual gifts sent subsequently by the audience user after the target virtual gifts are sent can be compared, the final virtual gift with the largest gift consumption resource is determined from the multiple target virtual gifts sent subsequently and the target virtual gift sent for the first time, and a record video is generated according to the gift display effect of the final virtual gift and the target live content, and the record video is sent to the audience user. The corresponding target live content in the record video can be the live content of the specified behavior for the final virtual gift that the audience user likes, can be the live content of the specified behavior for all target virtual gifts that the audience user likes, and can also be the live content of the regular performance behavior that the audience user likes.

[0046] In this step, the gift display effect of the target virtual gift and the target live content containing the positive feedback information of the audience user are generated to generate a record video, and the record video is automatically sent to the audience user, which can realize personalized editing for the audience user and improve the experience of the audience user.

[0047] The live video processing method, in response to the target virtual gift sent by the audience user in the live room, listens to the first feedback information of the audience user to the live content in the live room after sending the target virtual gift; in response to the positive feedback information in the first feedback information, the first live content corresponding to the positive feedback information in the first feedback information is determined as the target live content; based on the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user. In this way, after the audience user sends the target virtual gift in the live room, the live server listens to the feedback information of the audience user to the live content in the live room after sending the target virtual gift. After listening to the positive feedback information, the target live content corresponding to the positive feedback information is obtained. Finally, according to the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user. This way, by listening to the feedback information of the audience user to the live content in the live room after sending the target virtual gift, the gift display effect of the target virtual gift and the video clip of the live content liked by the audience user in the live video are automatically extracted, and a record video is generated and sent to the audience user. The automatic editing of the video clip of the gift display effect and the process of the anchor user expressing thanks to the audience user can be realized, the accuracy and quality of video extraction are improved, and the experience of the audience user is improved.

[0048] The following embodiments provide an implementation manner of listening to the feedback information of the live content in the live room.

[0049] In response to the target virtual gift sent by the audience user in the live room, whether the anchor user in the live room has a specified behavior for the target virtual gift is listened to; wherein the specified behavior includes a language thanksgiving behavior for the target virtual gift and / or a performance behavior; when it is determined that the anchor user has the specified behavior for the target virtual gift, a feedback interface for the specified behavior is provided to the audience user, and the feedback interface is used to receive first feedback information for the specified behavior.

[0050] The specified behavior can be a language thanksgiving behavior for the target virtual gift and a performance behavior for the target virtual gift, or only one of the language thanksgiving behavior for the target virtual gift or the performance behavior for the target virtual gift. The performance behavior for the target virtual gift includes singing performance for the target virtual gift, dance performance for the target virtual gift, etc.

[0051] When the live broadcast server recognizes that the audience user has sent the target virtual gift, it monitors whether the host user in the live broadcast room has performed a specified behavior in response to the target virtual gift. For example, by monitoring the host user's voice, if the host user mentions the account nickname of the audience user who sent the target virtual gift, the gift name of the target virtual gift, or key words such as "thank you" or "thank you" in the voice, it is determined that the host user has performed a verbal thank you behavior in response to the target virtual gift. For another example, when the live broadcast server monitors the host user's singing, dancing, and other performances, and these performances occur within a preset time period after the target virtual gift is sent, it is determined that the host user has performed a performance in response to the target virtual gift.

[0052] After determining that the host user has performed a specified behavior in response to the target virtual gift, the live broadcast server provides the viewer user with a feedback interface for the specified behavior and receives the viewer user's first feedback information regarding the specified behavior through the feedback interface. For example, while the host user is expressing their gratitude or performing a performance in response to the target virtual gift, a rating pop-up window will appear in the live broadcast room content interface displayed on the viewer user's terminal device. The viewer user can rate the host user's behavior within the provided rating range and based on their own preferences. The live broadcast server receives the score information given by the viewer user through the rating pop-up window and uses it as the feedback information regarding the specified behavior in the first feedback information.

[0053] The following embodiments provide implementation methods for determining target live broadcast content.

[0054] In one method, the target live broadcast content is a designated behavior that the audience likes and that occurs in response to a target virtual gift.

[0055] Specifically, in response to monitoring positive feedback information in the first feedback information of the user regarding the specified behavior through the feedback interface, the first live broadcast content in which the anchor user performs the specified behavior is determined as the target live broadcast content.

[0056] During or after the host user's designated actions, the live streaming server provides the audience user with a feedback interface for these designated actions. Through this feedback interface, the live streaming server can monitor and receive feedback from the audience user. When the live streaming server receives positive feedback from the feedback interface, it indicates that the audience user approves of and enjoys the host user's designated actions. The live streaming server then retrieves the host user's designated actions from the feedback interface and identifies the first live streaming content in which the designated actions occurred as the target live streaming content.

[0057] Specifically, in response to listening to positive feedback information in the first feedback information of the audience user for the specified behavior through the feedback interface, the specified behavior corresponding to the feedback interface is determined; the specified behavior is the behavior of the anchor user for the target virtual gift, and the specified behavior includes a plurality of sub-behaviors, each sub-behavior corresponding to a sub-behavior feedback information; the positive feedback information in the sub-behavior feedback information is the feedback information containing the specified feedback keyword; the target sub-behavior corresponding to the positive feedback information in the sub-behavior feedback information is determined from the specified behavior, and the first live content in which the target sub-behavior occurs is determined as the target live content.

[0058] The feedback information containing the specified feedback keyword is the feedback information containing positive evaluation, such as an evaluation score higher than a preset score threshold or an option name selected by the user. For example, for a feedback interface in the form of a scoring pop-up window, when the feedback information obtained by the live server is an evaluation score higher than a preset score threshold, the feedback information is determined as feedback information containing the specified feedback keyword. For a feedback interface in the form of an option pop-up window, when the live server receives that the audience user has selected the “satisfied” option, the feedback information obtained by the live server is feedback information containing the specified feedback keyword. The feedback information containing the specified feedback keyword is determined as the positive feedback information in the sub-behavior feedback information.

[0059] It can be understood that when the anchor user performs the specified behavior for the target virtual gift, the specified behavior includes a language thank-you behavior, a singing performance behavior for the target virtual gift, a dance performance behavior for the target virtual gift, and the like, the feedback interface provided by the live server displays each sub-behavior of the anchor user for the target virtual gift and a sub-behavior feedback information input area corresponding to each sub-behavior, the audience user inputs feedback information in each sub-behavior corresponding sub-behavior feedback information input area, and each sub-behavior has corresponding sub-behavior feedback information. When the live server listens to the positive feedback information in the sub-behavior feedback information through the feedback interface, the target sub-behavior corresponding to the positive feedback information in the sub-behavior feedback information is determined from the specified behavior, and the first live content in which the target sub-behavior occurs is determined as the target live content.

[0060] In another way, the anchor user in the live room does not perform the specified behavior for the target virtual gift, at this time, the target live content can be determined from the live content corresponding to all positive feedback information in the live session, that is, from all live content liked by the audience user in the live session.

[0061] The target live content can be determined by the following way:

[0062] In response to the second feedback information of the audience user for the live content in the live room, the second live content corresponding to the positive feedback information in the second feedback information is obtained; in response to the audience user sending the target virtual gift in the live room, it is listened whether the specified behavior of the anchor user in the live room occurs for the target virtual gift; wherein the specified behavior includes: language thanksgiving behavior and / or performance behavior for the target virtual gift; it is determined that the anchor user does not occur the specified behavior for the target virtual gift, and the target live content is determined from the second live content corresponding to the positive feedback information in the second feedback information.

[0063] The second feedback information is the feedback information of the audience user for the live content in the live room during the live process, and the second live content is the live content corresponding to the positive feedback information in the second feedback information.

[0064] That is, during the live process in the live room, the live server obtains each feedback information of the audience user for the live content in the live room, determines the second live content corresponding to the positive feedback information in the feedback information, and listens whether the specified behavior of the anchor user in the live room occurs for the target virtual gift after the audience user sends the target virtual gift in the live room. If it is determined that the anchor user does not occur the specified behavior for the target virtual gift, the target live content is determined from the live content corresponding to all positive feedback information of the audience user during the live process in the live room.

[0065] In this way, the positive feedback information in the second feedback information includes the specified bullet screen keyword, the bullet screen sending time of the bullet screen information containing the specified bullet screen keyword is determined, the live content corresponding to the anchor behavior occurring at the bullet screen sending time is determined as the second live content.

[0066] The specified bullet screen keyword refers to positive, positive, and praise words or phrases, such as "good" "brave" "like" and the like.

[0067] That is, the positive feedback information in the second feedback information can be the bullet screen information containing the specified bullet screen keyword, and the live content corresponding to the anchor behavior occurring at the bullet screen sending time is recorded according to the bullet screen sending time of the bullet screen information containing the specified bullet screen keyword, and is determined as the second live content.

[0068] In a specific implementation, the target live content can be the most favorite regular performance behavior of the audience user during the live process.

[0069] Specifically, the second live content contains a regular performance behavior of the anchor user; the regular performance behavior contained in the second live content is identified, for each regular performance behavior identified in the second live content, the start and end time of the regular performance behavior and the number of positive feedback information corresponding to the start and end time are determined, and the positive feedback information density is calculated; based on the density value corresponding to the positive feedback information density, the maximum positive feedback information density corresponding to the maximum density value is determined, and the regular performance behavior corresponding to the maximum positive feedback information density is obtained, and the live content of the regular performance behavior corresponding to the maximum positive feedback information density is determined as the target live content.

[0070] The above-mentioned regular performance behavior includes singing performance and dance performance of the anchor user, and also includes other program performance behaviors such as music performance, product introduction, etc. It should be noted that the singing performance and the dance performance in the regular performance behavior are different from the singing performance for the target virtual gift and the dance performance for the target virtual gift. The regular performance behavior is not a specified behavior of the anchor user after the target virtual gift is sent by the audience user. As an example, one way of judging is that the occurrence time of the regular performance behavior is not within a preset time period after the sending time point of the target virtual gift. The start and end time of the above-mentioned regular performance behavior refers to the start time point and the end time point of the regular performance behavior identified by the live server. The above-mentioned positive feedback information density can be understood as the number of positive feedback information per unit time within the duration of the regular performance behavior.

[0071] Here, the live server identifies the regular performance behavior contained in the second live content from the recorded second live content, determines the start and end time of the regular performance behavior and the number of positive feedback information corresponding to the start and end time for each regular performance behavior identified in the second live content, calculates the positive feedback information density, obtains the density value of the positive feedback information density corresponding to each regular performance behavior, finds the maximum positive feedback information density corresponding to the maximum value in the density value, obtains the regular performance behavior corresponding to the maximum positive feedback information density, and determines the live content of the regular performance behavior corresponding to the maximum positive feedback information density as the target live content. The target live content determined by this method is the regular performance behavior most liked by the audience user.

[0072] The following provides a way to calculate the positive feedback information density:

[0073] determine a start time point and an end time point of the regular performance behavior, and obtain an issuing time of the positive feedback information in the second feedback information, determine a number of information pieces of the positive feedback information whose issuing time is within the time interval of the start time point and the end time point from the second feedback information as the number of information pieces of the corresponding positive feedback information within the start time point and the end time point of the regular performance behavior, and calculate the positive feedback information density based on the duration time and the number of information pieces of the corresponding positive feedback information within the start time point and the end time point.

[0074] That is, a start time point and an end time point of the regular performance behavior are determined, the duration time of the regular performance behavior is obtained according to the start time point and the end time point, the issuing time of the positive feedback information in the second feedback information is obtained, the number of information pieces of the positive feedback information whose issuing time is within the start time point and the end time point of the regular performance behavior is determined from the second feedback information as the number of information pieces of the corresponding positive feedback information within the start time point and the end time point of the regular performance behavior, and the positive feedback information density is obtained by dividing the number of information pieces of the corresponding positive feedback information within the start time point and the end time point of the regular performance behavior by the duration time of the regular performance behavior.

[0075] The following embodiments provide an implementation manner of generating a record video.

[0076] In one manner, a first video segment containing a gift display effect of a target virtual gift is obtained, if the target live content contains language thanksgiving behavior of a host user for the target virtual gift and performance behavior for the target virtual gift, the start time and the end time of the language thanksgiving behavior and the performance behavior for the target virtual gift are determined, a second video segment containing the language thanksgiving behavior and a third video segment containing the performance behavior for the target virtual gift are generated, and a record video is generated based on the first video segment, the second video segment and the third video segment.

[0077] That is, a first video segment containing a gift display effect of a target virtual gift is obtained, if the target live content contains language thanksgiving behavior of a host user for the target virtual gift and performance behavior for the target virtual gift, the start time and the end time of the language thanksgiving behavior and the performance behavior for the target virtual gift are determined, a second video segment containing the language thanksgiving behavior and a third video segment containing the performance behavior for the target virtual gift are generated, and a record video is generated based on the first video segment, the second video segment and the third video segment.

[0078] Of course, if the target live content contains the language thanksgiving behavior of the anchor user for the target virtual gift or the performance behavior for the target virtual gift, the first video segment containing the gift display effect of the target virtual gift is acquired, and the live content in which the language thanksgiving behavior occurs or the live content in which the performance behavior for the target virtual gift occurs is extracted as the second video segment according to the start and end time of the language thanksgiving behavior or the start and end time of the performance behavior for the target virtual gift, and the recording video is generated according to the first video segment and the second video segment containing the regular performance behavior, and the recording video is sent to the audience user.

[0079] The above generation manner of the recording video can realize automatic clipping of the video segment of the gift display effect and the thanksgiving process of the anchor user to the audience user, and improves the accuracy and quality of video extraction.

[0080] In another manner, the anchor user in the live room does not have the specified behavior for the target virtual gift, in this case, the target live content does not contain the specified behavior for the target virtual gift, at this time, the target live content can be determined from the live content liked by the audience in the current live, in a specific implementation manner, the target live content can be the regular performance behavior most liked by the audience user in the current live, in this manner, the generation manner of the recording video is as follows:

[0081] Acquiring the first video segment containing the gift display effect of the target virtual gift; if the target live content contains the regular performance behavior of the anchor user, generating the second video segment containing the regular performance behavior based on the start and end time of the regular performance behavior; and generating the recording video based on the first video segment and the second video segment.

[0082] That is, the second video segment containing the most liked regular performance behavior is generated according to the start and end time of the most liked regular performance behavior contained in the target live content, and the recording video is generated according to the arrangement order of the first video segment and the second video segment together with the first video segment containing the gift display effect of the target virtual gift.

[0083] In this manner, if the anchor user in the live room does not have the specified behavior for the target virtual gift, the live server can also clip the regular performance behavior most liked by the audience user in the current live into the recording video, and send the recording video together with the first video segment containing the gift display effect of the target virtual gift to the audience user, thereby improving the experience of the audience user.

[0084] In another manner, for the behavior of sending the target virtual gift multiple times in the current live, the recording video is generated as follows:

[0085] In the current live broadcast in the live room, if the audience user sends multiple target virtual gifts, the final virtual gift is determined from the multiple target virtual gifts based on the gift consumption resources of the target virtual gifts; and the recording video is generated based on the gift display effect of the final virtual gift and the target live broadcast content.

[0086] That is, in the current live broadcast in the live room, if the audience user sends multiple target virtual gifts, the final virtual gift with the largest gift consumption resource can be determined from the multiple target virtual gifts sent subsequently and the target virtual gift sent for the first time by comparing the gift consumption resources of all target virtual gifts sent subsequently by the audience user after the target virtual gifts are sent; the recording video is generated according to the gift display effect of the final virtual gift and the target live broadcast content, and the recording video is sent to the audience user. The corresponding target live broadcast content in the recording video can be the live broadcast content of the specified behavior for the final virtual gift that the audience user likes, or the live broadcast content of the specified behavior for all target virtual gifts that the audience user likes, or the live broadcast content of the regular performance behavior that the audience user likes.

[0087] The following embodiments provide an implementation manner of sending the recording video to the audience user.

[0088] In response to a first specified event, the recording video is sent to the audience user; the first specified event includes the end of the current live broadcast or receiving a video sending instruction from the anchor user terminal; the live broadcast account information of the audience user is included in the video sending instruction.

[0089] In one way, after the current live broadcast ends, the live server will automatically send the recording video to the corresponding audience user terminal. In another way, the anchor user needs to touch the specified control on the terminal device interface of the anchor user to generate a video sending interface, and the anchor user can set the audience user individually or in batches in the video sending interface. After the anchor user touches the video sending control in the video sending interface, the terminal device of the anchor user generates a video sending instruction and sends it to the live server. The live broadcast account information of the audience user is included in the video sending instruction. After receiving the video sending instruction, the live server obtains the live broadcast account information of the audience user in the video sending instruction, determines the recording video corresponding to each audience user according to the live broadcast account information of the audience user, and automatically sends it to the corresponding audience user terminal.

[0090] In one specific embodiment, as shown in Figure 2 The processing method flow of the live video is introduced in detail.

[0091] In step S202, the live server identifies the target virtual gift and starts to listen to the first feedback information for the live broadcast content in the live room.

[0092] The live streaming server can identify the occurring anchor behaviors in the live streaming room and mark the start and end time of various behaviors in the live streaming room, and record the live streaming content.

[0093] Specifically, the sending time point of the target virtual gift is marked.

[0094] The live streaming server can identify the language thanksgiving behavior occurring for the target virtual gift, and record the start and end time points of the language thanksgiving behavior and the live streaming content. In one way, the live streaming server can identify the speech sent by the anchor user in real time, and when the account nickname of the audience user sending the target virtual gift, the gift name of the target virtual gift, or the keywords such as "thank you" and "thank you" are mentioned in the speech, it is determined that the anchor user has a language thanksgiving behavior for the target virtual gift. The live streaming server marks the time point when the sentence containing the keywords appears and the time point when the sentence ends, obtains the start and end time of the language thanksgiving behavior, records the live streaming content corresponding to the start and end time of the language thanksgiving behavior as the live streaming content of the anchor user's language thanksgiving behavior for the target virtual gift.

[0095] The live streaming server can also identify the singing performance behavior of the anchor user, and record the start and end time points of the corresponding behavior and the live streaming content. In one way, the voice of the anchor user in the live streaming room can be identified in real time according to the anchor's voice timbre, and when the song is identified, it is determined that the anchor user is singing. The live streaming server marks the time point when the anchor user starts singing and the time point when the anchor user stops singing, obtains the start and end time of the singing performance, and records the live streaming content corresponding to the start and end time of the singing performance as the live streaming content of the anchor user's singing performance behavior.

[0096] The live streaming server can also identify the dancing performance behavior of the anchor user. In one way, the live streaming platform can also use computer vision technology to analyze the video stream in the live streaming room, detect the anchor's actions and gestures in the live streaming content through real-time analysis of the video stream, and use a gesture estimation algorithm to detect whether the anchor is dancing. After determining that the anchor user is dancing, the live streaming server marks the time point when the anchor user starts dancing and the time point when the anchor user stops dancing, obtains the start and end time of the dancing performance, and records the live streaming content corresponding to the start and end time of the dancing performance as the live streaming content of the anchor user's dancing performance behavior. For example, when the live streaming server identifies that the anchor user's performance behavior such as singing and dancing occurs within a preset time period after the target virtual gift sending time point, it can be determined that the performance behavior is a performance behavior occurring for the target virtual gift.

[0097] If the anchor user performs the specified behavior in the live streaming room for the target virtual gift:

[0098] In one way, the live server can identify the specified behavior and provide the feedback interface for the thanksgiving behavior to the audience user. For example, when the live server identifies that the host user has performed the language thanksgiving behavior for the target virtual gift or the performance behavior for the target virtual gift, a scoring pop-up window is popped up in the live room content interface displayed on the terminal device of the audience user, and the audience user can score the various behaviors of the host within the provided scoring interval according to his own will. If the scoring interval is 1-5 points, the audience user scores 4 or 5 points for the performance behavior of the host, it is judged that the audience user likes the performance, and the live platform labels the performance with "xxx user favorite-thanksgiving performance" tag, and automatically cuts the live recording screen segment of this time period as the target live content.

[0099] In another way, the live user can identify through the barrage system that the audience user makes a positive evaluation on the specified behavior when the host performs the specified behavior for the target virtual gift, or the audience user continues to send gifts when the host performs the specified performance, and it can also be judged that the audience user likes the specified behavior, and the live platform also labels the specified behavior with "xxx user favorite-thanksgiving performance" tag, and automatically cuts the live recording screen segment of this time period as the target live content.

[0100] If the live server does not identify the thanksgiving behavior performed by the host user in the live room, the live server identifies all positive feedback information published by the audience user in the current live process, and records the corresponding label (number of keywords, time point of appearance). When the label time period of the positive feedback behavior of the barrage system and the host behavior forms a correspondence (for example, the host is performing a regular dance performance, and the live platform identifies that the audience user publishes a lot of praise type barrage), it is judged that the audience user likes the host behavior, and according to the positive feedback behavior, the audience user's favorite performance in the live room is sorted in the background. In the current live, this performance is labeled with "xxx user favorite-other performance", and the live recording screen segment of this time period is automatically cut as the target live content.

[0101] Step S204, judging whether the host user performs the specified behavior for the target virtual gift; if yes, entering step S206; if no, entering step S216;

[0102] The above-mentioned specified behavior can be a language thanksgiving behavior for the target virtual gift and a performance behavior for the target virtual gift, or only one of a language thanksgiving behavior for the target virtual gift or a performance behavior for the target virtual gift, and the performance behavior for the target virtual gift includes a singing performance for the target virtual gift and a dance performance for the target virtual gift.

[0103] When the account nickname of the audience user sending the target virtual gift, the gift name of the target virtual gift, or the keywords such as "thank you" and "thank you" are mentioned in the anchor voice, it is determined that the anchor user has the language thanksgiving behavior for the target virtual gift.

[0104] The live server identifies the voice of the anchor user in the live room according to the anchor voice, and performs real-time song recognition. When the song is recognized, it is determined that the anchor user is singing. The live server uses computer vision technology to analyze the video stream in the live room, detects the anchor's actions and gestures in the live content through real-time analysis of the video stream, and uses a gesture estimation algorithm to detect whether the anchor is dancing. When the live server identifies that the singing, dancing and other performance behaviors of the anchor user occur within the preset time period after the gift sending time point, it is determined that the anchor user has the performance behavior for the target virtual gift.

[0105] Step S206, determine whether the audience user likes the specified behavior for the target virtual gift; if yes, go to step S208; if no, go to step S210;

[0106] In one way, the live server can recognize the thanksgiving behavior and provide a feedback interface for the thanksgiving behavior to the audience user. For example, when the live server identifies that the anchor user performs singing and dancing performances for the target virtual gift, a scoring pop-up window is popped up in the live room content interface displayed on the terminal device of the audience user, and the audience user can score the specified behavior of the anchor within the provided scoring interval according to his own will. If the scoring interval is 1-5 points, the audience user scores 4 or 5 points for the dancing performance of the anchor, and it is determined that the audience user likes the dancing performance for the target virtual gift. If the audience user scores less than 4 points for the dancing performance of the anchor, it is determined that the audience user does not like the dancing performance.

[0107] In another way, the live server can recognize that the audience user makes positive comments on the anchor's performance when the anchor performs the specified behavior for the target virtual gift through the barrage system, or the audience user continues to send virtual gifts, and it can be determined that the audience user likes the specified behavior. If the audience user does not make positive comments on the anchor's performance when the anchor performs the specified behavior for the target virtual gift, it is determined that the audience user does not like the specified behavior.

[0108] Step S208, generate a recording video, and the recording video is a gift display effect for the target virtual gift.

[0109] Step S210, determine whether the audience user sends the target virtual gift multiple times within the live time; if no, go to step S212; if yes, go to step S214.

[0110] The live broadcast server can determine whether the viewer user has sent the target virtual gift multiple times based on the number of gift sending instructions received and the gift consumption resources in the gift sending instructions.

[0111] Step S212, taking the designated behavior that is popular with the audience as the target live broadcast content, generating a recorded video based on the gift display effect of the target virtual gift and the target live broadcast content, and determining the video template corresponding to the recorded video as: 1. The gift display effect of the target virtual gift; 2. The verbal thank you behavior for the target virtual gift; 3. The performance behavior for the target virtual gift.

[0112] Step S214, determine the final virtual gift, generate a recorded video based on the gift display effect of the final virtual gift and the target live broadcast content; determine the thank-you template corresponding to the recorded video as: 1. The gift display effect of the final virtual gift; 2. The verbal thank-you behavior for the final virtual gift; 3. All the live broadcast content that the audience user judges to be favorite (combined in chronological order).

[0113] Step S216: Identify whether the audience user has a favorite regular performance behavior in the current live broadcast through the bullet screen system; if not, proceed to step S218; if yes, proceed to step S220;

[0114] If the live broadcast server does not recognize that the anchor user has performed the specified behavior for the target virtual gift, the live broadcast server obtains all the barrage information sent by the audience user during the current live broadcast. If there is barrage information containing praise words, it is determined that the audience user has liked the live broadcast content in the current live broadcast; if there is no barrage information containing praise words, it is determined that there is no live broadcast content that the audience user likes in the current live broadcast.

[0115] Step S218: Generate a recorded video, and generate a recorded video as a gift display effect of the target virtual gift.

[0116] Step S220: Determine whether the audience user has sent the target virtual gift multiple times during the live broadcast of the current session; if not, proceed to step S222; if yes, proceed to step S224;

[0117] Step S222: The most favorite routine performance behavior of the audience users in the current live broadcast is used as the target live broadcast content. According to the gift display animation of the target virtual gift and the target live broadcast content, a recorded video is generated, and the video template corresponding to the recorded video is determined to be: 1. The gift display animation of the target virtual gift; 2. The most favorite routine performance behavior of the audience users in the current live broadcast.

[0118] Step S224, determine the final virtual gift, all host behaviors that are liked by the audience user as the target live content, generate a record video; determine the video template corresponding to the record video as: 1. The gift display effect of the final virtual gift; 2. All regular performance behaviors liked by the audience user (combined in chronological order).

[0119] After the host live broadcast is completed, the live video is automatically edited according to the video template corresponding to the record video, and is reorganized into a record video, which is sent to the user who sent the gift interaction through the host private message, and a script is matched "thank you for your XXX gift name, dance / song for you (judged according to the host behavior in the record video)".

[0120] In this way, after the live server recognizes that the audience user sends the target virtual gift, the recording of the live room content is automatically started, and the corresponding video clip is recorded. The live server obtains the feedback information of the audience user through the feedback interface, the bullet screen information and the gift sending instruction, determines the host behavior liked by the audience user, determines the record video through a series of logical judgments, matches different video templates for automatic video editing of the recorded video clip, matches the corresponding script, and automatically sends it to the user who interacts with the gift. The virtual gift display effect and the automatic editing of the host's expression of thanks process can be realized, the accuracy and quality of video extraction are improved, and if there is no host's expression of thanks process, the live server can also edit the regular performance behaviors liked by the audience user in the live broadcast into the record video and send it to the audience user, thereby improving the experience of the audience user.

[0121] Corresponding to the method embodiment described above, referring to Figure 3 The device includes:

[0122] The first monitoring module 302 is configured to, in response to the audience user sending the target virtual gift in the live room, monitor the first feedback information of the audience user to the live content in the live room after sending the target virtual gift.

[0123] The first determination module 304 is configured to, in response to monitoring the positive feedback information in the first feedback information, determine the first live content corresponding to the positive feedback information in the first feedback information as the target live content.

[0124] The first sending module 306 is configured to generate a record video based on the gift display effect of the target virtual gift and the target live content, and send the record video to the audience user.

[0125] In the manner, after the audience user sends out the target virtual gift in the live room, the live server listens to the feedback information of the audience user to the live content in the live room after sending out the target virtual gift. After listening to the positive feedback information, the target live content corresponding to the positive feedback information is obtained. Finally, according to the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user. Through listening to the feedback information of the audience user to the live content in the live room after sending out the target virtual gift, the gift display effect of the target virtual gift and the video clip of the live content liked by the audience user in the live video are automatically extracted, and the record video is generated and sent to the audience user. The automatic editing of the video clip of the gift display effect and the process of the anchor user expressing thanks to the audience user can be realized, the accuracy and quality of video extraction are improved, and the experience of the audience user is improved.

[0126] The first listening module is configured to listen to whether the anchor user in the live room has a specified behavior in response to the audience user sending out a target virtual gift in the live room. The specified behavior includes a verbal thanksgiving behavior and / or a performance behavior for the target virtual gift. When it is determined that the anchor user has the specified behavior for the target virtual gift, a feedback interface for the specified behavior is provided to the audience user. The feedback interface is configured to receive first feedback information for the specified behavior.

[0127] The first determination module is further configured to determine, in response to listening to positive feedback information in the first feedback information of the audience user for the specified behavior through the feedback interface, a first live content in which the anchor user has the specified behavior as the target live content.

[0128] The first determination module is further configured to determine, in response to listening to positive feedback information in the first feedback information of the audience user for the specified behavior through the feedback interface, the specified behavior corresponding to the feedback interface. The specified behavior is a behavior of the anchor user for the target virtual gift. The specified behavior includes a plurality of sub-behaviors, and each sub-behavior corresponds to a sub-behavior feedback information. The positive feedback information in the sub-behavior feedback information is feedback information containing a specified feedback keyword. The target sub-behavior corresponding to the positive feedback information in the sub-behavior feedback information is determined from the specified behavior, and the first live content in which the target sub-behavior occurs is determined as the target live content.

[0129] The apparatus further includes a second determination module configured to, in response to second feedback information of the audience user for the live content in the live room, obtain second live content corresponding to positive feedback information in the second feedback information; in response to the audience user sending a target virtual gift in the live room, monitor whether the anchor user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a verbal thanksgiving behavior and / or a performance behavior for the target virtual gift; and determine target live content from the second live content corresponding to the positive feedback information in the second feedback information, in response to the anchor user not having the specified behavior for the target virtual gift.

[0130] The apparatus further includes a third determination module configured to determine a sending time of the comment information containing the specified comment keyword, and determine live content corresponding to an anchor behavior occurring at the sending time of the comment information as the second live content.

[0131] The second live content contains a regular performance behavior of the anchor user; the apparatus further includes a fourth determination module configured to identify the regular performance behavior contained in the second live content, determine a start time and an end time of each regular performance behavior identified in the second live content and a number of positive feedback information in the start time and the end time, and calculate a positive feedback information density; determine a maximum positive feedback information density corresponding to a maximum value of a density value based on the density value corresponding to the positive feedback information density, and obtain a regular performance behavior corresponding to the maximum positive feedback information density, and determine live content of the regular performance behavior corresponding to the maximum positive feedback information density as the target live content.

[0132] The apparatus further includes a first calculation module configured to determine a start time and an end time of the regular performance behavior, calculate a duration corresponding to the regular performance behavior, obtain sending times of the positive feedback information in the second feedback information, determine a number of the positive feedback information in the second feedback information whose sending times are located in a time interval of the start time and the end time based on the start time and the end time and the sending times of the positive feedback information in the second feedback information, and determine the number of the positive feedback information as a number of the positive feedback information corresponding to the start time and the end time; and calculate the positive feedback information density based on the duration and the number of the positive feedback information corresponding to the start time and the end time.

[0133] The first sending module is further configured to obtain a first video clip containing a gift display effect of the target virtual gift; if the target live content contains a verbal thanksgiving behavior of the anchor user for the target virtual gift and a performance behavior for the target virtual gift, generate a second video clip containing the verbal thanksgiving behavior and a third video clip containing the performance behavior for the target virtual gift based on a start time and an end time of the verbal thanksgiving behavior and a start time and an end time of the performance behavior for the target virtual gift; and generate a record video based on the first video clip, the second video clip, and the third video clip.

[0134] The first sending module is further configured to acquire a first video clip containing a gift display dynamic effect of the target virtual gift; if the target live content contains a regular performance behavior of the anchor user, generate a second video clip containing the regular performance behavior based on start and end times of the regular performance behavior; and generate the recording video based on the first video clip and the second video clip.

[0135] The first sending module is further configured to, in the current live session in the live room, if the audience user sends multiple target virtual gifts, determine a final virtual gift from the multiple target virtual gifts based on gift consumption resources of the target virtual gifts; and generate the recording video based on a gift display dynamic effect of the final virtual gift and the target live content.

[0136] The first sending module is further configured to send the recording video to the audience user in response to a first specified event; the first specified event includes an end of the current live session or a video sending instruction received from the anchor user terminal; and the video sending instruction contains live account information of the audience user.

[0137] The embodiment also provides a live server, including a processor and a memory, the memory storing machine executable instructions capable of being executed by the processor, and the processor executes the machine executable instructions to implement the live video processing method.

[0138] Referring to Figure 4 The live server includes a processor 100 and a memory 101, the memory 101 storing machine executable instructions capable of being executed by the processor 100, and the processor 100 executes the machine executable instructions to implement the generation method of the outer contour stroke effect.

[0139] Further, Figure 4 The live server shown in the figure further includes a bus 102 and a communication interface 103, and the processor 100, the communication interface 103 and the memory 101 are connected through the bus 102. The memory 101 may contain a high-speed random access memory (RAM) and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 103 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network, etc. can be used. The bus 102 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 4Only one bidirectional arrow is used to represent multiple buses and multiple types of buses. The processor 100 can be an integrated circuit chip with signal processing capability. In implementation, each step of the above method can be completed by integrated logic circuits or instructions in the form of software in the processor 100. The processor 100 described above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium in the art. The storage medium is located in the memory 101, and the processor 100 reads the information in the memory 101, and combines the hardware to complete the steps of the method of the above embodiments.

[0140] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: in response to the target virtual gift sent by the audience user in the live room, listening to the first feedback information of the audience user after sending the target virtual gift to the live content in the live room; in response to the positive feedback information in the first feedback information, the first live content corresponding to the positive feedback information in the first feedback information is determined as the target live content; based on the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user.

[0141] In the manner, after the audience user sends out the target virtual gift in the live room, the live server listens to the feedback information of the audience user to the live content in the live room after sending out the target virtual gift. After listening to the positive feedback information, the target live content corresponding to the positive feedback information is obtained. Finally, according to the gift display effect of the target virtual gift and the target live content, a record video is generated, and the record video is sent to the audience user. Through listening to the feedback information of the audience user to the live content in the live room after sending out the target virtual gift, the gift display effect of the target virtual gift and the video clip of the live content liked by the audience user in the live video are automatically extracted, and the record video is generated and sent to the audience user. The automation of the video clip of the gift display effect and the process of the anchor user expressing thanks to the audience user can be realized, the accuracy and quality of the video extraction are improved, and the experience of the audience user is improved.

[0142] The processor in the live server can realize the following operations of the live video processing method by executing machine executable instructions: in response to the audience user sending out the target virtual gift in the live room, listening to whether the anchor user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a language thanksgiving behavior and / or a performance behavior for the target virtual gift; determining that the anchor user has the specified behavior for the target virtual gift, providing a feedback interface for the specified behavior to the audience user, and the feedback interface is used to receive first feedback information for the specified behavior.

[0143] The processor in the live server can realize the following operations of the live video processing method by executing machine executable instructions: in response to listening to the positive feedback information in the first feedback information of the audience user for the specified behavior through the feedback interface, determining the first live content in which the anchor user has the specified behavior as the target live content.

[0144] The processor in the live server can realize the following operations of the live video processing method by executing machine executable instructions: in response to listening to the positive feedback information in the first feedback information of the audience user for the specified behavior through the feedback interface, determining the specified behavior corresponding to the feedback interface; wherein the specified behavior is the behavior of the anchor user for the target virtual gift, the specified behavior includes a plurality of sub-behaviors, each sub-behavior corresponds to a sub-behavior feedback information; the positive feedback information in the sub-behavior feedback information is the feedback information containing the specified feedback keyword; the target sub-behavior corresponding to the positive feedback information in the sub-behavior feedback information is determined from the specified behavior, and the first live content in which the target sub-behavior occurs is determined as the target live content.

[0145] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: in response to second feedback information of the audience user for the live content in the live room, obtaining second live content corresponding to positive feedback information in the second feedback information; in response to the audience user sending a target virtual gift in the live room, listening to whether the anchor user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a verbal thanksgiving behavior and / or a performance behavior for the target virtual gift; determining that the anchor user does not have the specified behavior for the target virtual gift, and determining target live content from the second live content corresponding to the positive feedback information in the second feedback information.

[0146] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: determining a time of sending the comment information containing the specified comment keyword, and determining live content corresponding to the anchor behavior at the time of sending the comment as the second live content.

[0147] The second live content contains a regular performance behavior of the anchor user; the processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: recognizing the regular performance behavior contained in the second live content, determining the start and end time of the regular performance behavior and the number of corresponding positive feedback information in the start and end time for each regular performance behavior recognized in the second live content, and calculating the positive feedback information density; determining the maximum positive feedback information density corresponding to the maximum value of the density value based on the density value corresponding to the positive feedback information density, and obtaining the regular performance behavior corresponding to the maximum positive feedback information density, and determining the live content of the regular performance behavior corresponding to the maximum positive feedback information density as the target live content.

[0148] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: determining the start and end time of the regular performance behavior, and calculating the duration corresponding to the regular performance behavior; obtaining the sending time of the positive feedback information in the second feedback information; based on the start and end time and the sending time of the positive feedback information in the second feedback information, determining the number of information of the positive feedback information in the second feedback information whose sending time is in the time interval of the start and end time, and determining the number of information as the number of positive feedback information corresponding to the start and end time; based on the duration and the number of positive feedback information corresponding to the start and end time, calculating the positive feedback information density.

[0149] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: obtaining a first video clip containing a gift display dynamic effect of a target virtual gift; if the target live content contains a language thanksgiving behavior of the anchor user for the target virtual gift and a performance behavior for the target virtual gift, generating a second video clip containing the language thanksgiving behavior and a third video clip containing the performance behavior for the target virtual gift based on the start and end time of the language thanksgiving behavior and the start and end time of the performance behavior for the target virtual gift; and generating a recording video based on the first video clip, the second video clip and the third video clip.

[0150] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: obtaining a first video clip containing a gift display dynamic effect of a target virtual gift; if the target live content contains a regular performance behavior of the anchor user, generating a second video clip containing the regular performance behavior based on the start and end time of the regular performance behavior; and generating a recording video based on the first video clip and the second video clip containing the regular performance behavior.

[0151] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: if multiple target virtual gifts are sent by the audience user in the current session live in the live room, determining a final virtual gift from the multiple target virtual gifts based on gift consumption resources of the target virtual gifts; and generating a recording video based on a gift display dynamic effect of the final virtual gift and the target live content.

[0152] The processor in the live server can implement the following operations of the live video processing method by executing machine executable instructions: sending the recording video to the audience user in response to a first specified event; the first specified event includes an end of the current session live or receiving a video sending instruction from the anchor user terminal; and the video sending instruction contains live account information of the audience user.

[0153] The embodiment also provides a machine readable storage medium, which stores machine executable instructions, and the machine executable instructions cause the processor to implement the generation method of the outer contour stroke effect when the machine executable instructions are called and executed by the processor.

[0154] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: in response to the target virtual gift sent by the audience user in the live room, listening to the first feedback information of the audience user after sending the target virtual gift to the live content in the live room; in response to the positive feedback information in the first feedback information, determining the first live content corresponding to the positive feedback information in the first feedback information as the target live content; generating a record video based on the gift display effect of the target virtual gift and the target live content, and sending the record video to the audience user.

[0155] In this way, after the audience user sends the target virtual gift in the live room, the live server listens to the feedback information of the audience user after sending the target virtual gift to the live content in the live room, obtains the target live content corresponding to the positive feedback information after listening to the positive feedback information, and finally generates a record video based on the gift display effect of the target virtual gift and the target live content, and sends the record video to the audience user. This way automatically extracts the gift display effect of the target virtual gift and the video clip of the live content liked by the audience user in the live video by listening to the feedback information of the audience user after sending the target virtual gift to the live content in the live room, and generates a record video and sends it to the audience user, which can realize automatic editing of the video clip of the gift display effect and the process of the anchor user expressing thanks to the audience user, improve the accuracy and quality of video extraction, and improve the experience of the audience user.

[0156] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: in response to the target virtual gift sent by the audience user in the live room, listening to whether the anchor user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a language thanksgiving behavior and / or a performance behavior for the target virtual gift; determining that the anchor user has the specified behavior for the target virtual gift, providing a feedback interface for the specified behavior to the audience user, and the feedback interface is used to receive the first feedback information for the specified behavior.

[0157] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: in response to the target virtual gift sent by the audience user in the live room, listening to whether the anchor user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a language thanksgiving behavior and / or a performance behavior for the target virtual gift; determining that the anchor user has the specified behavior for the target virtual gift, providing a feedback interface for the specified behavior to the audience user, and the feedback interface is used to receive the first feedback information for the specified behavior.

[0158] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: in response to the positive feedback information in the first feedback information of the audience user for the specified behavior, determining the specified behavior corresponding to the feedback interface; wherein the specified behavior is the behavior of the host user for the target virtual gift, and the specified behavior includes a plurality of sub behaviors, and each sub behavior corresponds to a sub behavior feedback information; the positive feedback information in the sub behavior feedback information is the feedback information containing the specified feedback keyword; determining the target sub behavior corresponding to the positive feedback information in the sub behavior feedback information from the specified behavior, and determining the first live content in which the target sub behavior occurs as the target live content.

[0159] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: in response to the second feedback information of the audience user for the live content in the live room, obtaining the second live content corresponding to the positive feedback information in the second feedback information; in response to the audience user sending the target virtual gift in the live room, listening to whether the host user in the live room occurs the specified behavior for the target virtual gift; wherein the specified behavior includes: language thanksgiving behavior and / or performance behavior for the target virtual gift; determining that the host user does not occur the specified behavior for the target virtual gift, and determining the target live content from the second live content corresponding to the positive feedback information in the second feedback information.

[0160] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: determining the time of sending the specified barrage keyword containing the barrage information, and determining the live content corresponding to the host behavior occurring at the time of sending the barrage as the second live content.

[0161] The second live content contains the regular performance behavior of the host user; the machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: identifying the regular performance behavior contained in the second live content, determining the start and end time of the regular performance behavior and the number of corresponding positive feedback information in the start and end time for each regular performance behavior identified in the second live content, and calculating the positive feedback information density; based on the density value corresponding to the positive feedback information density, determining the maximum positive feedback information density corresponding to the maximum value of the density value, obtaining the regular performance behavior corresponding to the maximum positive feedback information density, and determining the live content of the regular performance behavior corresponding to the maximum positive feedback information density as the target live content.

[0162] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: determining the start time and the end time of the regular performance behavior, calculating the duration corresponding to the regular performance behavior; obtaining the sending time of the positive feedback information in the second feedback information; determining the information number of the positive feedback information in the second feedback information based on the start time and the end time and the sending time of the positive feedback information in the second feedback information, and determining the information number as the corresponding positive feedback information number in the start time and the end time; calculating the positive feedback information density based on the duration and the corresponding positive feedback information number in the start time and the end time.

[0163] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: obtaining a first video segment containing a gift display effect of a target virtual gift; if the target live content contains a language thanksgiving behavior of the anchor user for the target virtual gift and a performance behavior for the target virtual gift, generating a second video segment containing the language thanksgiving behavior and a third video segment containing the performance behavior for the target virtual gift based on the start time and the end time of the language thanksgiving behavior and the start time and the end time of the performance behavior for the target virtual gift; and generating a record video based on the first video segment, the second video segment and the third video segment.

[0164] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: obtaining a first video segment containing a gift display effect of a target virtual gift; if the target live content contains a regular performance behavior of the anchor user, generating a second video segment containing the regular performance behavior based on the start time and the end time of the regular performance behavior; and generating a record video based on the first video segment and the second video segment containing the regular performance behavior.

[0165] The machine executable instructions stored in the machine readable storage medium can realize the following operation in the live video processing method: if multiple target virtual gifts are sent by the audience user in the current live session in the live room, determining a final virtual gift from the multiple target virtual gifts based on the gift consumption resource of the target virtual gift; and generating a record video based on the gift display effect of the final virtual gift and the target live content.

[0166] The machine-executable instructions stored in the machine-readable storage medium described above, by executing the machine-executable instructions, can implement the following operation in the live video processing method described above: in response to a first specified event, sending the recorded video to the audience user; the first specified event includes: an end of a current live or a video sending instruction received from the anchor user terminal; the video sending instruction contains live account information of the audience user.

[0167] The live video processing method, device, live server and computer program product of the storage medium provided by the embodiments of the present application include a computer readable storage medium storing program codes, the instructions included in the program codes can be used to execute the method described in the foregoing method embodiments, and the specific implementation can be referred to the method embodiments, which will not be described here.

[0168] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0169] In addition, in the description of the embodiments of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connection" and "connection" should be understood in a broad sense, for example, can be fixedly connected, can be detachably connected, or integrally connected; can be mechanically connected, or electrically connected; can be directly connected, or indirectly connected through an intermediate medium, can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0170] The functions, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the part of the present application which essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk and various program code storage media.

[0171] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first", "second", "third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0172] Finally, it should be noted that the above embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, and are not limiting. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical range disclosed by the present application, or make equivalent replacements to some technical features; and these modifications, changes or replacements do not cause the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A live video processing method, characterized in that, The method comprises: in response to a target virtual gift sent by a viewer user in a live room, listening to first feedback information of the viewer user for live content in the live room after sending the target virtual gift; in response to listening to positive feedback information in the first feedback information, determining first live content corresponding to the positive feedback information in the first feedback information as target live content; based on the gift display effect of the target virtual gift and the target live content, generating a recording video, and sending the recording video to the viewer user.

2. The method of claim 1, wherein, The step of listening to first feedback information of a viewer user for live content in a live room in response to the viewer user sending a target virtual gift in the live room comprises: in response to a target virtual gift sent by a viewer user in a live room, listening to whether a host user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a verbal thanksgiving behavior and / or a performance behavior for the target virtual gift; determining that the host user has a specified behavior for the target virtual gift, providing a feedback interface for the specified behavior to the viewer user, and the feedback interface is used to receive the first feedback information for the specified behavior.

3. The method of claim 2, wherein, The step of determining first live content corresponding to the positive feedback information in the first feedback information as target live content in response to listening to positive feedback information in the first feedback information comprises: in response to listening to positive feedback information in the first feedback information of the viewer user for the specified behavior through the feedback interface, determining first live content in which the host user has the specified behavior as target live content.

4. The method of claim 3, wherein, The step of determining first live content corresponding to the positive feedback information in the first feedback information as target live content in response to listening to positive feedback information in the first feedback information comprises: in response to listening to positive feedback information in the first feedback information of the viewer user for the specified behavior through the feedback interface, determining the specified behavior corresponding to the feedback interface; wherein the specified behavior is a behavior of the host user for the target virtual gift, and the specified behavior includes a plurality of sub-behaviors, each of which corresponds to a sub-behavior feedback information; positive feedback information in the sub-behavior feedback information is feedback information containing a specified feedback keyword; determining a target sub-behavior corresponding to positive feedback information in the sub-behavior feedback information from the specified behavior, and determining first live content in which the target sub-behavior occurs as target live content.

5. The method of claim 1, wherein, The method further comprises: in response to second feedback information of a viewer user for live content in a live room, obtaining second live content corresponding to positive feedback information in the second feedback information; in response to a target virtual gift sent by a viewer user in a live room, listening to whether a host user in the live room has a specified behavior for the target virtual gift; wherein the specified behavior includes a verbal thanksgiving behavior and / or a performance behavior for the target virtual gift; It is determined that the specified behavior does not occur to the target virtual gift of the anchor user, and target live content is determined from the second live content corresponding to the positive feedback information in the second feedback information.

6. The method of claim 5, wherein, The positive feedback information in the second feedback information includes a specified bullet screen keyword containing bullet screen information; the step of obtaining the second live content corresponding to the positive feedback information in the second feedback information comprises: Determine the bullet screen sending time of the bullet screen information containing the specified bullet screen keyword, and determine the live content corresponding to the anchor behavior occurring at the bullet screen sending time as the second live content.

7. The method of claim 5, wherein, The second live content contains the regular performance behavior of the anchor user; The step of determining the target live content from the second live content corresponding to the positive feedback information in the second feedback information comprises: Identify the regular performance behavior contained in the second live content, determine the start and end time of each regular performance behavior identified in the second live content and the number of corresponding positive feedback information within the start and end time, and calculate the positive feedback information density; Based on the density value corresponding to the positive feedback information density, determine the maximum positive feedback information density corresponding to the maximum value of the density value, and obtain the regular performance behavior corresponding to the maximum positive feedback information density, and determine the live content occurring in the regular performance behavior corresponding to the maximum positive feedback information density as the target live content.

8. The method of claim 7, wherein, The step of determining the start and end time of the regular performance behavior and the number of corresponding positive feedback information within the start and end time to calculate the positive feedback information density comprises: Determine the start and end time of the regular performance behavior, and calculate the duration corresponding to the regular performance behavior; Obtain the sending time of the positive feedback information in the second feedback information; Based on the start and end time and the sending time of the positive feedback information in the second feedback information, determine the information number of the positive feedback information in the second feedback information whose sending time is located in the time interval of the start and end time, and determine the information number as the number of corresponding positive feedback information within the start and end time; Based on the duration and the number of corresponding positive feedback information within the start and end time, calculate the positive feedback information density.

9. The method of claim 1, wherein, Based on the gift display effect of the target virtual gift and the target live content, the step of generating a record video comprises: Obtain a first video segment containing the gift display effect of the target virtual gift; If the target live content contains the language thanksgiving behavior of the anchor user for the target virtual gift and the performance behavior for the target virtual gift, based on the start and end time of the language thanksgiving behavior and the start and end time of the performance behavior for the target virtual gift; generate a second video segment containing the language thanksgiving behavior, and a third video segment containing the performance behavior for the target virtual gift; Generate a record video based on the first video segment, the second video segment and the third video segment.

10. The method of claim 1 or 5, wherein, Based on the gift display effect of the target virtual gift and the target live content, the step of generating a record video further comprises: obtaining a first video clip containing a gift display effect of the target virtual gift; if the target live content contains a regular performance behavior of the anchor user, generating a second video clip containing the regular performance behavior based on start and end time of the regular performance behavior; generating a record video based on the first video clip and the second video clip containing the regular performance behavior.

11. The method of claim 1, wherein, The step of generating a record video based on the target virtual gift display effect and the target live content further comprises: if the audience user sends multiple target virtual gifts in the current live session, determining a final virtual gift from the multiple target virtual gifts based on gift consumption resources of the target virtual gifts; generating a record video based on the final virtual gift display effect and the target live content.

12. The method of claim 1, wherein, The step of sending the record video to the audience user comprises: sending the record video to the audience user in response to a first specified event; the first specified event includes the end of the current live session or receiving a video sending instruction from the anchor user terminal; the video sending instruction contains live account information of the audience user.

13. A live video processing apparatus, characterized by comprising: The device is arranged in a live server, and the device comprises: a first listening module, which listens to first feedback information of the audience user for live content in the live room after sending the target virtual gift in response to the audience user sending the target virtual gift; a first determination module, which determines first live content corresponding to positive feedback information in the first feedback information as target live content in response to listening to the positive feedback information in the first feedback information; a first sending module, which generates a record video based on a gift display effect of the target virtual gift and the target live content, and sends the record video to the audience user.

14. A live server, characterized by The machine-readable storage medium stores machine-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the live video processing method of any one of claims 1-12.

15. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores machine-executable instructions, which, when invoked and executed by a processor, cause the processor to implement the live video processing method of any one of claims 1-12.

Images