CN115914690A

CN115914690A - Data processing method, device, server and storage medium

Info

Publication number: CN115914690A
Application number: CN202211348169.7A
Authority: CN
Inventors: 杜海涛
Original assignee: Beijing Dajia Internet Information Technology Co Ltd
Current assignee: Beijing Dajia Internet Information Technology Co Ltd
Priority date: 2022-10-31
Filing date: 2022-10-31
Publication date: 2023-04-04

Abstract

The disclosure relates to a data processing method, an apparatus, a server and a storage medium. The method comprises the following steps: acquiring audio and video stream fragments of a live broadcast room; determining a live broadcast state of the live broadcast room according to the audio and video stream segments, wherein the live broadcast state comprises an output state of a user live broadcast interaction behavior of the live broadcast room; and adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level. The embodiment of the disclosure can effectively reduce the occurrence of recommending live broadcasting rooms, which are not output by the user live broadcasting interaction behavior, to the user, and improve the accuracy of recommendation.

Description

Data processing method, device, server and storage medium

Technical Field

The present disclosure relates to the field of internet technologies, and in particular, to a data processing method, an apparatus, a server, and a storage medium.

Background

With the development of internet technology and streaming media, live video technology has emerged. The video live broadcast is gradually a mainstream expression mode of the internet due to the combination of rich elements such as images, characters and sounds and the like and the combination of sound and form. However, some live rooms without a live user often appear due to reasons such as the live room anchor temporarily leaving or forgetting to end the live. At this time, if the live broadcast room is recommended to the audience, the recommendation effect is greatly reduced.

Disclosure of Invention

The present disclosure provides a data processing method, apparatus, server and storage medium to at least solve the problem of poor recommendation effect in the related art. The technical scheme of the disclosure is as follows:

according to a first aspect of the embodiments of the present disclosure, there is provided a data processing method, including:

acquiring audio and video stream fragments of a live broadcast room;

determining a live broadcast state of the live broadcast room according to the audio and video stream segments, wherein the live broadcast state comprises an output state of a user live broadcast interaction behavior of the live broadcast room;

and adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior of the live broadcast room is not output, and the audio/video stream segment includes a plurality of audio/video stream sub-segments which are periodically acquired in the live broadcast room; the determining the live broadcasting state of the live broadcasting room according to the audio and video stream clip comprises the following steps:

determining a first initial live broadcast state of the live broadcast room corresponding to each audio and video stream sub-segment according to the plurality of audio and video stream sub-segments;

and under the condition that the continuous accumulated value of the occurrence times of the first initial live broadcast state is greater than the preset times, determining that the live broadcast state of the live broadcast room is the first state.

In a possible implementation manner, the determining, according to a plurality of audio/video stream sub-segments, a first initial live broadcasting state of the live broadcasting room corresponding to each audio/video stream sub-segment includes:

respectively extracting an image sub-segment and an audio sub-segment corresponding to each audio-video stream sub-segment from a plurality of audio-video stream sub-segments;

determining a second initial live broadcast state of the live broadcast room according to the image sub-segment, and determining a third initial live broadcast state of the live broadcast room according to the audio sub-segment;

determining a joint initial live broadcast state of the live broadcast room according to the second initial live broadcast state and the third initial live broadcast state;

the determining that the live broadcast state of the live broadcast room is the first state under the condition that the continuous accumulated value of the occurrence times of the first initial live broadcast state is the first state is greater than the preset times includes:

and under the condition that the continuous accumulated value of the occurrence times of the joint initial live broadcast state in the first state is greater than the preset times, determining that the live broadcast state of the live broadcast room is in the first state.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior in the live broadcast room is not output, and the audio/video stream segment includes a continuous audio/video stream segment within a preset time period; the determining the live broadcasting state of the live broadcasting room according to the audio and video stream clip comprises the following steps:

and under the condition that the audio and video stream fragments are not updated in the preset time period, determining that the live broadcast state of the live broadcast room is a first state.

In a possible implementation manner, determining that a live broadcast state of the live broadcast room is a first state when the audio/video stream segment is not updated in the preset time period includes:

extracting audio clips and image clips from the audio and video clips;

and under the condition that the audio clip and the image clip are not updated in the preset time period, determining that the live broadcasting state of the live broadcasting room is a first state.

In a possible implementation manner, the outputting state includes a first state in which a user live broadcast interaction behavior of the live broadcast room is not output, the adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and performing a corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level includes:

under the condition that the live broadcast state of the live broadcast room is a first state, sending verification information to the live broadcast room by using a long-connection communication mode;

and under the condition that the feedback information corresponding to the verification information is not received within a preset time length, reducing the recommendation level of the live broadcast room, and recommending the live broadcast room according to the reduced recommendation level.

In one possible implementation, the output state includes a second state and a third state; the second state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is greater than or equal to a preset score value; the third state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is smaller than a preset score value; the method for adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room and performing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level comprises the following steps:

under the condition that the live broadcast state of the live broadcast room is in a second state, improving the recommendation level of the live broadcast room, and recommending the live broadcast room according to the improved recommendation level;

and under the condition that the live broadcast state of the live broadcast room is in a third state, reducing the recommendation level of the live broadcast room, and recommending the live broadcast room according to the reduced recommendation level.

In a possible implementation manner, the acquiring an audio/video stream segment of a live broadcast includes:

acquiring an access address of an audio and video stream segment of a live broadcast room from a message queue, wherein the access address is a storage address generated after a content distribution network server intercepts the audio and video stream segment of each live broadcast room;

and accessing a content distribution network server by using the access address, and acquiring the audio and video stream segments of the live broadcast room from the content distribution network server.

According to a second aspect of the embodiments of the present disclosure, there is provided a data processing apparatus including:

the acquisition module is used for acquiring audio and video stream clips of a live broadcast room;

the determining module is used for determining the live broadcasting state of the live broadcasting room according to the audio and video stream fragments, wherein the live broadcasting state comprises the output state of the user live broadcasting interaction behavior of the live broadcasting room;

and the recommending module is used for adjusting the recommending level of the live broadcast room according to the live broadcast state of the live broadcast room and executing corresponding recommending operation on the live broadcast room according to the adjusted recommending level.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior of the live broadcast room is not output, and the audio/video stream segment includes a plurality of audio/video stream sub-segments which are periodically acquired in the live broadcast room; the determining module comprises:

the first determining submodule is used for determining a first initial live broadcast state of the live broadcast room corresponding to each audio and video stream sub-segment according to the plurality of audio and video stream sub-segments;

and the second determining submodule is used for determining that the live broadcast state of the live broadcast room is the first state under the condition that the continuous accumulated value of the occurrence times of the first initial live broadcast state is greater than the preset times.

In one possible implementation, the first determining sub-module includes:

the first extraction unit is used for respectively extracting an image sub-segment and an audio sub-segment corresponding to each audio and video stream sub-segment from a plurality of audio and video stream sub-segments;

the first determining unit is used for determining a second initial live broadcasting state of the live broadcasting room according to the image sub-segment and determining a third initial live broadcasting state of the live broadcasting room according to the audio sub-segment;

a second determining unit, configured to determine, according to the second initial live broadcast state and the third initial live broadcast state, a joint initial live broadcast state of the live broadcast room;

the second determination submodule includes:

and a third determining unit, configured to determine that the live broadcast state of the live broadcast room is the first state when the continuous accumulated value of the occurrence times of the joint initial live broadcast state being the first state is greater than a preset time.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior in the live broadcast room is not output, and the audio/video stream segment includes a continuous audio/video stream segment within a preset time period; the determining module comprises:

and the third determining submodule is used for determining that the live broadcast state of the live broadcast room is the first state under the condition that the audio and video stream fragments are not updated in the preset time period.

In one possible implementation manner, the third determining submodule includes:

the second extraction unit is used for extracting audio clips and image clips from the audio and video clips;

and the fourth determining unit is used for determining that the live broadcast state of the live broadcast room is the first state under the condition that the audio clip and the image clip are not updated in the preset time period.

In one possible implementation manner, the output state includes a first state in which a user live interaction behavior of the live broadcast room is not output, and the processing module includes:

the sending submodule is used for sending verification information to the live broadcast room by using a long-connection communication mode under the condition that the live broadcast state of the live broadcast room is a first state;

and the adjusting submodule is used for reducing the recommendation level of the live broadcast room under the condition that the feedback information corresponding to the verification information is not received within the preset duration, and recommending the live broadcast room according to the reduced recommendation level.

In one possible implementation, the output state includes a second state and a third state; the second state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is greater than or equal to a preset score value; the third state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is smaller than a preset score value; the processing module comprises:

the improvement submodule is used for improving the recommendation level of the live broadcast room under the condition that the live broadcast state of the live broadcast room is in a second state, and recommending the live broadcast room according to the improved recommendation level;

and the reduction submodule is used for reducing the recommendation level of the live broadcast room under the condition that the live broadcast state of the live broadcast room is a third state, and recommending the live broadcast room according to the reduced recommendation level.

In one possible implementation manner, the obtaining module includes:

the first acquisition submodule is used for acquiring an access address of an audio/video stream segment of a live broadcast room from the message queue, wherein the access address is a storage address generated after the content distribution network server intercepts the audio/video stream segment of each live broadcast room;

and the second acquisition submodule is used for accessing the content distribution network server by using the access address and acquiring the audio and video stream fragments of the live broadcast room from the content distribution network server.

According to a third aspect of the embodiments of the present disclosure, there is provided a server, including:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the data processing method according to any one of the embodiments of the present disclosure.

According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, wherein instructions of the computer-readable storage medium, when executed by a processor of a server, enable the server to perform the data processing method according to any one of the embodiments of the present disclosure.

According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product comprising instructions which, when executed by a processor of a server, enable the server to perform the data processing method according to any one of the embodiments of the present disclosure.

The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects: according to the embodiment of the disclosure, the live broadcasting state of the live broadcasting room is determined by automatically acquiring the audio and video stream segments of the live broadcasting room, and the recommendation level of the live broadcasting room is adjusted according to the live broadcasting state of the live broadcasting room, so that the occurrence that the live broadcasting room recommended to the user is not output by the user live broadcasting interaction behavior of the live broadcasting room is effectively reduced, and the recommendation accuracy is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.

FIG. 1 is a flow chart illustrating a method of data processing in accordance with an exemplary embodiment.

Fig. 2 is a schematic diagram illustrating a sliding window statistical live state in accordance with an exemplary embodiment.

Fig. 3 is a flow chart illustrating a method of data processing according to another exemplary embodiment.

Fig. 4 is a flow chart illustrating a method of data processing according to another exemplary embodiment.

FIG. 5 is a block diagram illustrating a data processing apparatus according to an example embodiment.

FIG. 6 is a block diagram illustrating a server in accordance with an example embodiment.

Detailed Description

In order to make the technical solutions of the present disclosure better understood, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

It should also be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for presentation, analyzed data, etc.) referred to in the present disclosure are both information and data that are authorized by the user or sufficiently authorized by various parties.

Fig. 1 is a flow chart illustrating a method of data processing, as shown in fig. 1, for use in a server, according to an exemplary embodiment, including the following steps.

And step S101, acquiring audio and video stream fragments of a live broadcast room.

In the embodiment of the present disclosure, the live broadcast room may include a real-time virtual room with network interaction, and the scenes of the live broadcast room may include various scenes, such as a video live broadcast room, a voice live broadcast room, a chat room live broadcast room, a singing live broadcast room, a game live broadcast room, a multi-main-broadcast competition live broadcast room, and the like. In the live broadcast room, a host uses a client to broadcast some content in a live broadcast mode, and the content in the form of sound, images, characters and the like is often accompanied in the live broadcast room. The viewer may view through a website or application. In the embodiment of the present disclosure, the audio/video stream segment may include a segment of live content in a preset time interval, for example, t ₁ To t ₂ The live content in between, such as a live video clip, may also include live content corresponding to a plurality of time intervals, such as t ₁ To t ₂ Live content, t in between ₃ To t ₄ Live content and t between ₅ To t ₆ Live content in between, etc. Of course, the corresponding audio/video stream segments may also include segments containing only audio, or segments containing only images, andincluding segments that are both audio and image.

And step S103, determining the live broadcast state of the live broadcast room according to the audio and video stream fragments, wherein the live broadcast state comprises the output state of the user live broadcast interaction behavior of the live broadcast room.

In this embodiment of the disclosure, the output state of the live broadcast interaction behavior may include a state where the user live broadcast interaction behavior in the live broadcast room is not output, and a state where the output quality is greater than or equal to a preset score value, and a state where the user live broadcast interaction behavior in the live broadcast room is output, and the output quality is less than a preset scoring system. The user of the live broadcast room may include a creator (or a main broadcast) of the live broadcast room, and may also include other users who are in the live broadcast room and enter by invitation of the creator, or enter actively, and the like. In one example, a live view of an audio-visual stream segment may show a user, such as a performing user; or when the picture or audio content of the user who is performing video teaching is updated, the user in the live broadcast room broadcasts the output state of the interactive behavior. In another example, the live view of the audio-video stream segment may also not show the user, for example, some scene views are shown, and at this time, if the scene is not updated and only the audio is updated, the status of the live interactive behavior output of the user in the live view may also be shown. Therefore, whether the user is displayed in the picture of the audio and video stream clip or not, the live broadcast room can output the state of the user live broadcast interaction behavior of the live broadcast room.

In the embodiment of the disclosure, the live broadcast interactive behavior output state of the user in the live broadcast room may include live broadcast interactive behavior output of the user in the live broadcast room, and the output quality is greater than or equal to the preset score value state and the live broadcast interactive behavior output of the user in the live broadcast room, and the output quality is less than the preset score value state. The output quality is used for representing the output quality of a live broadcast picture, and the output quality can be measured through a preset regular grading system and the like, for example, when a display user speaks in the live broadcast picture, but no corresponding sound exists in an audio and video stream segment, or the sound is not matched with an image, or actions in the picture are not smooth, the display is not clear, and the like, the output quality of the audio and video stream segment can be considered to be poor; correspondingly, for example, the output quality of the audio/video stream segment with smooth motion is better, the image of the picture is clear, the image and the sound are matched, and the output quality of the audio/video stream segment with smooth motion is better. The state that the live broadcast interaction behavior of the user in the live broadcast room is not output may include that the live broadcast content is not updated any more, for example, in a preset time period, the live broadcast content is not updated in any of a plurality of forms, where the plurality of forms specifically may include that an image is not updated, a sound is not updated, and a character is not updated.

In the embodiment of the disclosure, the live broadcast state of the live broadcast room is determined according to the audio and video stream fragments. The method specifically comprises the following steps: in one example, a plurality of audio and video stream segments are periodically acquired, and the live state in a live room is determined. E.g. t ₁ Time interval audio-video stream fragments a, t ₂ Time interval audio-video stream fragments b and t ₃ And if the audio and video stream segment a, the audio and video stream segment b and the audio and video stream segment c in the time interval all indicate that the live broadcast interaction behavior of the user in the live broadcast room is not output, indicating that the live broadcast state of the live broadcast room is the live broadcast interaction behavior of the user in the live broadcast room is not output. In another example, the audiovisual stream segment may include a segment of live content, e.g., t, acquired within a predetermined time interval ₁ To t ₅ And comparing whether the live broadcast content is updated or not in the time interval, and if the live broadcast content is not updated, determining that the live broadcast state of the live broadcast room is the live broadcast interaction behavior of the user in the live broadcast room and is not output. Specifically, when the content of the audio/video stream clip is judged whether to be updated, the live broadcast state of the live broadcast room can be comprehensively judged by combining multiple aspects of contents such as images, sounds and characters, which is not limited by the present disclosure. It should be noted that the manner of whether the image is updated or not may be implemented by various methods such as an image processing method, edge detection, an artificial intelligence network-based target identification method, and the like. The way of comparing whether the sound is updated can be realized by various methods such as voiceprint recognition, volume and tone detection methods, and the disclosure does not limit the methods.

And S105, adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level.

In the embodiment of the disclosure, when the live broadcast state is a state in which the live broadcast interaction behavior of the user in the live broadcast room is not output or the live broadcast interaction behavior of the user in the live broadcast room is output, and the output quality is less than the preset score value, the recommendation level corresponding to the live broadcast room is reduced, that is, the recommendation processing work of the live broadcast room is reduced. In one example, recommendation ranks of live rooms, such as live room a, live room b, live room c, live room d and live room e \8230, can be established according to recommendation levels, and when the situation that the live room c meets the condition of the un-output state or the output quality is lower than a preset score value is detected, the recommendation ranks of the live room c can be dropped, such as being placed behind the live room d. In another example, recommendations may be made in conjunction with a recommendation policy for the live room, such as a recommendation score for the live room equal to a weighted value in each recommendation dimension, for example: and when the live broadcasting room is detected to be in accordance with the un-output state or the output quality is smaller than the preset score value, the weight a corresponding to the X can be reduced to reduce the recommendation score of the live broadcasting room. Where Y and Z may represent other recommendation dimensions, such as viewer user preferences, search terms for viewer content, and so forth, which are not intended to be limiting of this disclosure. On the other hand, when the live broadcast state is the live broadcast interaction behavior output of the user in the live broadcast room, and the output quality is greater than or equal to the preset score value, the recommendation level corresponding to the live broadcast room is improved, for example, the ranking is improved or the weight value is improved in the above embodiment, which is not limited by the present disclosure.

According to the embodiment of the method and the device, the audio and video stream segments of the live broadcast room are automatically acquired to determine the live broadcast state of the live broadcast room, and the recommendation level of the live broadcast room is adjusted according to the live broadcast state of the live broadcast room, so that the occurrence that the live broadcast interactive behavior of the user in the live broadcast room is not output and recommended to the user in the live broadcast room is effectively reduced, and the recommendation accuracy is improved. On the other hand, the live broadcast room with the live broadcast interaction behavior not output is not directly recommended, the situations that no live broadcast room is recommended or the number of recommended live broadcast rooms is too small can be reduced, and the recommendation effect is improved.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior of the live broadcast room is not output, and the audio/video stream segment includes a plurality of audio/video stream sub-segments which are periodically acquired in the live broadcast room; the determining the live broadcast state of the live broadcast room according to the audio and video stream fragments comprises the following steps:

In the embodiment of the present disclosure, the plurality of audio/video stream sub-segments of the live broadcast room that are periodically obtained may include audio/video stream sub-segments that are obtained at preset intervals, for example, the audio/video stream sub-segments are obtained at intervals of 1 minute. In the embodiment of the present disclosure, each of the plurality of audio/video stream sub-segments may be detected, and a first initial live state corresponding to each of the audio/video stream sub-segments is determined. For example, detecting that a first initial live state of a first audio and video stream sub-segment is a first state; the first initial live broadcast state of the second audio and video stream sub-segment is not the first state; the first initial live state of the third audiovisual stream sub-segment is the first state. In the embodiment of the present disclosure, the preset times may include multiple times, such as 3 times, 5 times, 8 times, and the like, which is not limited by the present disclosure. In the embodiment of the present disclosure, the continuous accumulated value may include a continuous value of a live status corresponding to a sub-segment of the audio-video stream that is continuous in time, for example, t ₁ A first initial live broadcast state corresponding to a first audio and video stream sub-segment acquired at any moment is a first state; t is t ₂ A first initial live broadcast state corresponding to a second audio and video stream sub-segment acquired at any moment is not a first state; t is t ₃ A first initial live broadcast state corresponding to a third audio-video stream sub-segment acquired at any momentThe state is a first state; t is t ₄ A first initial live broadcast state corresponding to a fourth audio and video stream sub-segment acquired at any moment is a first state; t is t ₅ And a first initial live broadcast state corresponding to the fifth audio-video stream sub-segment acquired at the moment is a first state. For t ₃ At the time, the number of times of occurrence of the first state is continuously integrated into 2 times, i.e. t ₂ And t ₃ The time corresponds to a first state; for t ₄ At the time, the number of occurrences of the first state is continuously integrated into 3 times, i.e., t ₂ 、t ₃ 、t ₄ The time corresponds to a first state; for t ₅ At time, the continuous integration value of the first state is 4 times, i.e. t ₂ 、t ₃ 、t ₄ 、t ₅ The time corresponds to the first state. If the preset number of times is 3, t ₅ The continuous accumulated value of the occurrence times of the first state at the moment is larger than the preset times, and then the live broadcast room is determined to be in t ₅ The live state at the moment is the first state.

According to the embodiment of the present disclosure, the live broadcast state of the live broadcast room is determined by determining the occurrence frequency continuous accumulated value of the first initial live broadcast state as the first state, and comparing the accumulated value with the preset frequency. The method and the device can reduce the occurrence of the situation that the live broadcast state of the live broadcast room is not in line with the reality due to accidental errors of a certain audio and video stream sub-segment, and improve the accuracy of the judgment result.

In a possible implementation manner, determining, according to a plurality of audio/video stream sub-segments, a first initial live broadcast state of the live broadcast room corresponding to each of the audio/video stream sub-segments includes:

the determining of the first initial live broadcast state of the live broadcast room corresponding to each audio and video stream sub-segment according to the plurality of audio and video stream sub-segments includes:

In the embodiment of the present disclosure, the image sub-segment and the audio sub-segment may be comprehensively identified, for example, the image sub-segment and the audio sub-segment are respectively extracted from the audio/video stream sub-segment. Referring to fig. 2, the recognition result of the image sub-segment, that is, the second initial live broadcast state is stored in the memory, for example, the second initial live broadcast state stored in the array a [0] is output as the user live broadcast interaction behavior of the live broadcast room; the second initial live broadcast state stored in the array a [1] is that the user live broadcast interaction behavior of the live broadcast room is not output; the second initial live broadcast state stored in the array a [2] is the output of the user live broadcast interaction behavior of the live broadcast room, and the like. Storing the recognition result of the audio sub-segment, namely the third initial live broadcast state into a memory, wherein the second initial live broadcast state stored by the array b [0] is not output in the live broadcast interaction behavior of the user in a live broadcast room; the second initial live broadcast state stored in the array b [1] is output as the user live broadcast interaction behavior of a live broadcast room; the second initial live broadcast state stored in the array b [2] is that the user live broadcast interaction behavior of the live broadcast room is not output, and the like. Wherein, in fig. 2, "there" indicates the user live broadcast interactive behavior output of the live broadcast room; "none" indicates that the user live interaction behavior of the live room is not output.

In an embodiment of the present disclosure, the joint initial live state represents a result determined comprehensively according to the second initial live state of the image and the third initial live state of the audio. In one example, when the second initial live broadcast state is that the user live broadcast interaction behavior of the live broadcast room is not output, and the third initial live broadcast state is that the user live broadcast interaction behavior of the live broadcast room is not output, the user live broadcast interaction behavior of the joint initial live broadcast state of the live broadcast room is not output. In another example, where either or both of the second initial live state or the third initial live state is output as a user live interaction behavior of the live room, the joint initial live state is determined to be output as a user live interaction behavior of the live room. Referring to fig. 2, a joint initial live broadcast state is obtained according to the image recognition result and the audio recognition result. And under the condition that the continuous accumulated value of the occurrence times of the joint initial live broadcast state in the first state is greater than the preset times, determining that the live broadcast state of the live broadcast room is in the first state.

In one example, the joint initial live state may be counted in a sliding window manner. The sliding window comprises a calculation mode based on double pointers, and a window is formed between elements pointed by the two pointers. In fig. 2, the size of the window 201 is related to the setting of the preset number, for example, the preset number is 2, and the minimum case exceeding 2 is 3, the window size is set to 3 arrays. When the window 201 moves to the right, the first state appears three times in succession, indicating that the live state of the live room is the first state. In the embodiment of the present disclosure, the moving direction of the sliding window is not limited, and in another embodiment, the window 201 may also move leftward.

According to the embodiment of the disclosure, the accuracy of the identification result is improved by integrating the second initial live broadcast state of the image sub-segment and the third initial live broadcast state of the audio sub-segment, and the real-time performance of the judgment result is improved by judging the joint initial live broadcast state in a sliding window manner, so that the judgment result is output at every moment.

In the disclosed embodiment, the preset time period may be set, for example, 3 minutes or 10 minutes. In the embodiment of the present disclosure, the condition that the audio/video stream segment is not updated may include that the audio is not updated, the image is not updated, or the text is not updated. In the embodiment of the present disclosure, whether the audio/video segment is updated may be determined according to any one of the audio, the image, or the text, or whether the audio/video segment is updated may be determined according to any two or all of the audio, the image, or the text.

The method for judging the live broadcast state of the live broadcast room is simple and convenient to implement.

extracting audio clips and image clips from the audio and video clips;

and under the condition that the audio clip and the image clip are not updated in the preset time period, determining that the live broadcast state of the live broadcast room is a first state.

In the embodiment of the disclosure, the image segment and the audio segment can be respectively obtained from the audio/video stream segment; the image segment and the audio segment may also be obtained separately, for example, the audio segment is obtained by using a recording method, and the image segment is obtained by using a screenshot method, which is not limited in this disclosure. The mode of determining the live broadcast state of the live broadcast room according to the image segments can be realized by various methods such as an image processing method, an edge detection method, an artificial intelligence network-based target identification method and the like in the embodiment. According to the audio clip, the mode for determining the live broadcast state of the live broadcast room can be realized by various methods such as voiceprint recognition, volume and tone detection methods, and the like, which are not limited by the disclosure. There may be a correspondence between the image segment and the audio segment. Here, the correspondence indicates a correspondence in time, e.g., t ₁ Image segment and t acquired at moment ₁ Time of day is obtainedThe audio clip taken.

According to the embodiment of the disclosure, under the condition that the audio clip and the image clip are dually identified as the first state, the live broadcasting state of the live broadcasting room is determined to be the first state. The accuracy of the judgment result is improved, and the occurrence of inaccuracy of the judgment result possibly caused by the independent use of information of one dimension is reduced.

In the embodiment of the present disclosure, the verification information may include various forms, such as a verification code, a puzzle control, a confirmation control, and the like. And when the live broadcast state is the first state, sending verification information to the live broadcast room. And if the feedback information corresponding to the verification information is not received within a preset time period, for example, the feedback information is not received within 1 minute, reducing the recommendation level corresponding to the live broadcast room. In the embodiment of the present disclosure, the communication mode of the long connection is compared with the communication mode of the short connection, for example, in the long connection communication, the client initiates a connection to the server, and the server establishes a connection in a dual-sending manner after receiving the request. The client sends a message to the server, the server responds to the client, one-time reading and writing is completed, the connection is not closed, and the client initiates a closing request after long-time operation. And the communication mode of the short connection closes the connection after completing one-time reading and writing. In one embodiment, when the anchor establishes the live broadcast room, broadcast information can be sent to the server in a long-connection communication mode, the server periodically acquires audio and video stream segments of the live broadcast room, and when the audio and video stream segments are detected to be in a first state, verification information is sent to the live broadcast room.

According to the embodiment of the disclosure, by adding the verification information and the feedback information, the accuracy of the judgment of the first state can be enhanced, the misjudgment of the live broadcast room for the first state through the misjudgment of the audio and video stream fragments is prevented, the accuracy of the judgment result is improved, and the recommendation effect of the live broadcast room is improved. According to the embodiment of the disclosure, the verification information is sent to the live broadcast room through the long-connection communication mode, the client communication address corresponding to the live broadcast room does not need to be recorded, and the verification information is quickly and conveniently sent to the live broadcast room.

In one possible implementation, the output state includes a second state and a third state; the second state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is greater than or equal to a preset score value; the third state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is smaller than a preset score value; the adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level comprises the following steps:

In the embodiment of the present disclosure, the output quality is used for representing the output quality of the live broadcast picture, and the output quality can be measured by a preset rule of a scoring mechanism, and the third state can be expressed as that when a display user speaks in the live broadcast picture but no corresponding sound exists in an audio/video stream segment, or the sound is not matched with an image, or actions in the picture are not smooth, the display is not clear, and the like, at this time, the output quality of the audio/video stream segment can be considered to be poor, and the score is low; correspondingly, the second state can be expressed as that the image of the picture is clear, the image is matched with the sound, the output quality of the audio/video stream segment with smooth action is better, and the score is higher. In the embodiment of the present disclosure, the method for improving the recommendation level of the live broadcast room and the method for reducing the recommendation level of the live broadcast room may include any one of the above embodiments, and details of the present disclosure are not repeated here.

According to the embodiment of the disclosure, the state of the user live broadcast interaction behavior output in the live broadcast room is specifically divided into the second state and the third state, so that the second state and the third state are more in line with the actual situation, and a more flexible implementation mode is provided.

and accessing a content distribution network server by using the access address, and acquiring the audio and video stream fragments of the live broadcast room from the content distribution network server.

In the embodiment of the present disclosure, the Content Delivery Network (CDN) server may be configured to cache image, video, or audio data for a user to access. The content distribution network server can periodically intercept audio and video stream fragments of each live broadcast room, generate a storage address and intercept messages. Because the number of live broadcast rooms is large, in one example, the intercepted messages can be stored in the message queue according to the sequence of the generation time of the intercepted messages, wherein the intercepted messages include storage addresses. In one example, an access address of the audio and video stream segment of the live broadcast room may be obtained from the message queue, and the audio and video stream segment of the live broadcast room may be obtained according to the access address.

The embodiment of the disclosure provides an obtaining mode for obtaining audio and video stream clips of each live broadcast room when more live broadcast rooms are faced, and the judgment of live broadcast states of a plurality of live broadcast rooms can be performed in order.

Fig. 3 is a flow chart illustrating a method of data processing according to another exemplary embodiment. Referring to fig. 3, the method includes:

step S301, a plurality of audio and video stream sub-segments in a live broadcast room are periodically acquired.

In the embodiment of the present disclosure, the plurality of audio/video stream sub-segments of the live broadcast that are periodically obtained may include audio/video stream sub-segments that are obtained at preset time intervals, for example, the audio/video stream sub-segments are obtained at 1 minute intervals.

Step S303, respectively extracting an image sub-segment and an audio sub-segment corresponding to each audio/video stream sub-segment from the plurality of audio/video stream sub-segments.

Step S305, determining a second initial live broadcast state of the live broadcast room according to the image sub-segment, and determining a third initial live broadcast state of the live broadcast room according to the audio sub-segment.

And step S307, under the condition that the continuous accumulated value of the occurrence times of the joint initial live broadcast state in the first state is greater than the preset times, determining that the live broadcast state of the live broadcast room is in the first state.

In an embodiment of the present disclosure, the joint initial live broadcast state represents a result determined comprehensively according to the second initial live broadcast state of the image and the third initial live broadcast state of the audio. In one example, when the second initial live broadcast state is that the user live broadcast interaction behavior of the live broadcast room is not output, and the third initial live broadcast state is that the user live broadcast interaction behavior of the live broadcast room is not output, the user live broadcast interaction behavior of the joint initial live broadcast state of the live broadcast room is not output. In another example, where either or both of the second initial live state or the third initial live state is output as a user live interaction behavior of the live room, the joint initial live state is determined to be output as a user live interaction behavior of the live room.

Step S309, under the condition that the continuous accumulated value of the occurrence times of the joint initial live broadcast state in the first state is larger than the preset times, determining that the live broadcast state of the live broadcast room is in the first state.

In the embodiment of the present disclosure, the joint initial live broadcast state may be counted in a sliding window manner. The sliding window comprises a calculation mode based on double pointers, and a window is formed between elements pointed by the two pointers. In fig. 2, the size of the window 201 is related to the setting of the preset times, for example, the preset times are 2 times, and the minimum case exceeding 2 is 3 times, then the window size is set to 3 arrays. When the window 201 moves to the right, the first state appears three times in succession, indicating that the live state of the live room is the first state.

Step S311, when the live broadcast status of the live broadcast room is the first status, sending verification information to the live broadcast room by using a long-connection communication method.

In the embodiment of the present disclosure, the communication mode of the long connection is compared with the communication mode of the short connection, for example, in the long connection communication, the client initiates a connection to the server, and the server establishes a connection in a dual-sending manner after receiving the request. The client sends a message to the server, the server responds to the client, one-time reading and writing is completed, the connection is not closed, and the client initiates a closing request after long-time operation. And the communication mode of the short connection closes the connection after completing one-time reading and writing.

In the embodiment of the present disclosure, the verification information may include various forms, such as a verification code, a control of a sliding window, a confirmation control, and the like.

And step S313, under the condition that the feedback information corresponding to the verification information is not received within the preset time length, reducing the recommendation level of the live broadcast room, and recommending the live broadcast room according to the reduced recommendation level.

And when the live broadcast state is the first state, sending verification information to the live broadcast room. And if the feedback information corresponding to the verification information is not received within a preset time period, for example, the feedback information is not received within 1 minute, reducing the recommendation level corresponding to the live broadcast room.

Fig. 4 is a flow chart illustrating a method of data processing according to another exemplary embodiment. Referring to fig. 4, a live API (application program interface) module, a live task module for image sound detection, a long connection module, an MMU (memory management unit) module, an MMU information module, a screenshot task module, a screenshot and voice message module, a CDN module, and a database module all belong to a server side, and a module two words are omitted in the figure.

In fig. 4, where the anchor is located at the client, the steps of the method may include: 1. broadcasting: acquiring a broadcast message of a main broadcast in a live broadcast API module; 2. writing the broadcast information: writing the broadcast information into a database; 3. live pictures and sounds were captured, once in 1 minute: the CDN module intercepts live broadcast image clips and live broadcast audio clips of a live broadcast room every minute; 4. sending a slicing message: the screenshot task module sends screenshot messages to a screenshot and sound message module, wherein the screenshot messages comprise access addresses of audio and video stream clips, and a plurality of screenshot messages form a message queue in a screenshot and sound message model; 5. identify pictures and sounds, whether someone (user): the MMU module respectively identifies live broadcast image clips and live broadcast audio clips through an algorithm and judges the live broadcast state of a live broadcast room; 6. sending the recognition result of the picture and the sound: the identification result can comprise a live broadcast state, namely the live broadcast interaction behavior of the user in the live broadcast room is output or not output; 7. judging by using a sliding window, and comprehensively judging sound and picture results: and synthesizing the voice recognition result and the image recognition result, wherein when the two recognition results are both live broadcasts and no main broadcast exists in the live broadcast room, the recognition result is determined to be the absence of the main broadcast. The sliding window is used for counting the continuous accumulated value of the occurrence times without the anchor, and a judgment result is obtained at each moment, so that the real-time performance of judgment is improved; 8. writing a person identifier (live state of the output interaction behavior): if the user live broadcast interaction behavior of the live broadcast room is comprehensively judged to be output, the person identification is written into the database; 9.Skip: when the time is less than 3 times, the method indicates that audio and video stream fragments are continuously and periodically obtained, and the accumulated values of the obtained audio and video stream fragments and the audio and video stream fragments in the previous two times are counted; 10. and signaling is sent: if the judgment result is that no live broadcast state exists, sending a signaling to the long connection module; 11. sending the required verification code: the long connection module sends a verification code to a live broadcast room of the client; 12. writing the transmitted verification code identification: writing the transmitted verification code identification into a database to prevent repeated transmission; 13. submitting a verification code result: submitting feedback information in the live broadcast room; 14. checking the result of the verification code and checking whether the feedback information is correct or not; 15. and returning a verification result: returning a check result of the feedback information to the live broadcast room; 16. if the feedback information is received and the feedback information is correct, writing an identifier of the output interaction behavior of the live broadcast room into the database, and if the feedback information is not received within 1 minute; 17 mark as unmanned live room: and writing the identification of the interaction behavior which is not output in the live broadcast room into the data.

According to the embodiment of the present disclosure, the identifier of the live broadcast state is written into the database, which is beneficial for determining the recommendation level of the live broadcast room conveniently when the subsequent live broadcast room recommendation is performed. According to the method and the device, the audio and video stream fragments are periodically acquired in real time, the accuracy and the real-time performance of judgment of the live broadcast room are improved, the recommendation level of the live broadcast room is determined according to the live broadcast state, and the recommendation effect is improved.

It should be understood that, although the steps in the flowchart are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least some of the steps in the figures may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternatively with other steps or at least some of the steps or stages in other steps.

It is understood that the same/similar parts between the embodiments of the method described above in this specification can be referred to each other, and each embodiment focuses on the differences from the other embodiments, and it is sufficient that the relevant points are referred to the descriptions of the other method embodiments.

Based on the same inventive concept, the embodiment of the present disclosure further provides a target account identification apparatus for implementing the above-mentioned target account identification method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so the specific limitations in the following embodiments of the target account identification device can be referred to the limitations on the target account identification method in the above, and are not described herein again.

The apparatus may include systems (including distributed systems), software (applications), modules, components, servers, clients, etc. that use the methods described in embodiments of the present specification in conjunction with any necessary apparatus to implement the hardware. Based on the same innovative concept, the embodiments of the present disclosure provide an apparatus in one or more embodiments as described in the following embodiments. Since the implementation scheme of the apparatus for solving the problem is similar to that of the method, the specific implementation of the apparatus in the embodiment of the present specification may refer to the implementation of the foregoing method, and repeated details are not repeated. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

FIG. 5 illustrates a data processing apparatus according to an example embodiment. Referring to fig. 5, the apparatus 500 includes:

an obtaining module 501, configured to obtain an audio/video stream segment of a live broadcast room;

a determining module 503, configured to determine a live broadcast state of the live broadcast room according to the audio/video stream segment, where the live broadcast state includes an output state of a user live broadcast interaction behavior of the live broadcast room;

and the recommending module 505 is configured to adjust the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and execute a corresponding recommending operation on the live broadcast room according to the adjusted recommendation level.

In one possible implementation, the first determining sub-module includes:

the first determining unit is used for determining a second initial live broadcast state of the live broadcast room according to the image sub-segment and determining a third initial live broadcast state of the live broadcast room according to the audio sub-segment;

the second determination submodule includes:

and the third determining unit is used for determining that the live broadcast state of the live broadcast room is the first state under the condition that the continuous accumulated value of the occurrence times of the joint initial live broadcast state in the first state is greater than the preset times.

In a possible implementation manner, the output state includes a first state in which a user live broadcast interaction behavior of the live broadcast room is not output, and the audio/video stream segment includes continuous audio/video stream segments within a preset time period; the determining module comprises:

and the third determining submodule is used for determining that the live broadcast state of the live broadcast room is the first state under the condition that the audio and video stream clips are not updated in the preset time period.

In one possible implementation manner, the output state includes a first state in which a user live interaction behavior of the live room is not output, and the processing module includes:

and the adjusting submodule is used for reducing the recommendation level of the live broadcast room under the condition that the feedback information corresponding to the verification information is not received within the preset time length, and recommending the live broadcast room according to the reduced recommendation level.

In one possible implementation manner, the obtaining module includes:

the first obtaining submodule is used for obtaining an access address of an audio and video stream segment of a live broadcast room from a message queue, wherein the access address is a storage address generated after a content distribution network server intercepts the audio and video stream segment of each live broadcast room;

and the second acquisition submodule is used for accessing a content distribution network server by using the access address and acquiring the audio and video stream fragments of the live broadcast room from the content distribution network server.

With regard to the apparatus in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.

Fig. 6 is a block diagram illustrating an electronic device 600 for a data processing method according to an example embodiment. For example, the electronic device 600 may be a server. Referring to fig. 6, electronic device 600 includes a processing component 620 that further includes one or more processors, and memory resources, represented by memory 622, for storing instructions, such as application programs, that are executable by processing component 620. The application programs stored in memory 622 may include one or more modules that each correspond to a set of instructions. Further, the processing component 620 is configured to execute instructions to perform the above-described methods.

The electronic device 600 may further include: the power component 624 is configured to perform power management for the electronic device 600, the wired or wireless network interface 626 is configured to connect the electronic device 600 to a network, and the input/output (I/O) interface 628. The electronic device 600 may operate based on an operating system stored in the memory 622, such as Windows Server, mac OS X, unix, linux, freeBSD, or the like.

In an exemplary embodiment, a computer-readable storage medium comprising instructions, such as the memory 622 comprising instructions executable by a processor of the electronic device 600 to perform the above-described method is also provided. The storage medium may be a computer-readable storage medium, which may be, for example, a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product is also provided, which includes instructions executable by a processor of the electronic device 600 to perform the above-described method.

It should be noted that the descriptions of the above-mentioned apparatus, the electronic device, the computer-readable storage medium, the computer program product, and the like according to the method embodiments may also include other embodiments, and specific implementations may refer to the descriptions of the related method embodiments, which are not described in detail herein.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method of data processing, comprising:

acquiring audio and video stream fragments of a live broadcast room;

determining a live broadcast state of the live broadcast room according to the audio and video stream fragments, wherein the live broadcast state comprises an output state of a user live broadcast interaction behavior of the live broadcast room;

2. The method as claimed in claim 1, wherein the output state includes a first state in which a user live interaction behavior of the live broadcast room is not output, and the audio/video stream segment includes a plurality of audio/video stream sub-segments of the live broadcast room which are periodically acquired; the determining the live broadcast state of the live broadcast room according to the audio and video stream fragments comprises the following steps:

3. The method according to claim 2, wherein the determining a first initial live state of the live broadcast room corresponding to each of the audio/video stream sub-segments according to the plurality of audio/video stream sub-segments comprises:

respectively extracting an image sub-segment and an audio sub-segment corresponding to each audio and video stream sub-segment from a plurality of audio and video stream sub-segments;

the determining that the live broadcast state of the live broadcast room is the first state when the continuous accumulated value of the occurrence times of the first initial live broadcast state being the first state is greater than the preset times comprises:

4. The method according to claim 1, wherein the output state comprises a first state in which a user live interaction behavior of the live broadcast room is not output, and the audio/video stream segment comprises a continuous audio/video stream segment within a preset time period; the determining the live broadcast state of the live broadcast room according to the audio and video stream fragments comprises the following steps:

5. The method according to claim 4, wherein determining that the live broadcast state of the live broadcast room is the first state in a case that the audio/video stream segment is not updated within the preset time period comprises:

extracting audio clips and image clips from the audio and video clips;

6. The method as claimed in claim 1, wherein the output state includes a first state in which a user live interaction behavior of the live broadcast room is not output, the adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and performing a corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level includes:

7. The method of claim 1, wherein the output state comprises a second state and a third state; the second state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is greater than or equal to a preset score value; the third state is a state that the user live broadcast interaction behavior of the live broadcast room is output, and the output quality is smaller than a preset score value; the adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room, and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level comprises the following steps:

and under the condition that the live broadcasting state of the live broadcasting room is in a third state, reducing the recommendation level of the live broadcasting room, and recommending the live broadcasting room according to the reduced recommendation level.

8. The method according to any one of claims 1 to 7, wherein the acquiring of the live-air audio and video stream segment comprises:

9. A data processing apparatus, comprising:

the determining module is used for determining the live broadcast state of the live broadcast room according to the audio and video stream fragments, wherein the live broadcast state comprises the output state of the user live broadcast interaction behavior of the live broadcast room;

and the processing module is used for adjusting the recommendation level of the live broadcast room according to the live broadcast state of the live broadcast room and executing corresponding recommendation operation on the live broadcast room according to the adjusted recommendation level.

10. A server, comprising:

a processor;

a memory for storing the processor-executable instructions;

wherein the processor is configured to execute the instructions to implement the data processing method of any of claims 1 to 8.

11. A computer-readable storage medium, wherein instructions in the computer-readable storage medium, when executed by a processor of a server, enable the server to perform the data processing method of any of claims 1 to 8.