CN113438492A

CN113438492A - Topic generation method and system in live broadcast, computer equipment and storage medium

Info

Publication number: CN113438492A
Application number: CN202110614322.5A
Authority: CN
Inventors: 曾衍
Original assignee: Guangzhou Cubesili Information Technology Co Ltd
Current assignee: Guangzhou Cubesili Information Technology Co Ltd
Priority date: 2021-06-02
Filing date: 2021-06-02
Publication date: 2021-09-24
Anticipated expiration: 2041-06-02
Also published as: CN113438492B

Abstract

The application relates to a title generation method, a title generation system, computer equipment and a storage medium in live broadcast, wherein the method comprises the following steps: the method comprises the steps that a main broadcast client responds to content identification starting operation to obtain live broadcast audio and video data, the content data are identified from the live broadcast audio and video data, a question generation request is generated according to the content data, and the question generation request is sent to a server; the server responds to a title generation request of the anchor client, obtains content data, extracts keywords in the content data according to the content data and a preset semantic analysis algorithm, then obtains target titles matched with the keywords, and issues the target titles to the anchor client; and the anchor client receives the target title matched by the server according to the keywords in the content data and displays the target title. Compared with the prior art, the method and the device have the advantages that the titles can be quickly generated in the live broadcasting process for being used by the anchor, the title generation efficiency is improved, and the interactive experience between audiences and the anchor is effectively improved.

Description

Topic generation method and system in live broadcast, computer equipment and storage medium

Technical Field

The embodiment of the application relates to the technical field of network live broadcast, in particular to a title generation method, a title generation system, computer equipment and a storage medium in live broadcast.

Background

With the progress of network communication technology, live webcast becomes a new network interaction mode, and live webcast is popular with more and more audiences due to the characteristics of instantaneity, interactivity and the like.

Currently, during live webcasting, an online anchor often needs to interact with the audience, for example: giving virtual gifts, answering and leaving messages, and the like. For answering interaction, a director is often required to temporarily put out some questions or prepare some questions in advance.

If some questions are temporarily proposed, the time is relatively consumed, and the interactive experience of the user is influenced; if some questions are prepared in advance, the anchor cannot predict the current live broadcast condition, which may cause the questions not to meet the requirements of the audience, and also reduce the effect of question answering interaction.

Disclosure of Invention

The embodiment of the application provides a title generation method, a title generation system, computer equipment and a storage medium in live broadcast, which can solve the technical problems that the time consumption efficiency of title setting in the network live broadcast process is low and the interactive experience of a user is influenced, and the technical scheme is as follows:

in a first aspect, an embodiment of the present application provides a title generation method in live broadcasting, including the steps of:

the anchor client responds to the content identification starting operation to acquire live broadcast audio and video data; identifying content data from the live audio and video data; generating a title generation request according to the content data, and sending the title generation request to a server;

the server responds to a title generation request of the anchor client and acquires content data identified from live audio and video data of the anchor client; extracting key words in the content data according to the content data and a preset semantic analysis algorithm; obtaining a target title matched with the keyword according to the keyword; issuing the target title to the anchor client;

and the anchor client receives a target title matched by the server according to the keywords in the content data and displays the target title.

In a second aspect, an embodiment of the present application provides a title generation method in live broadcasting, including the steps of:

responding to a title generation request of a anchor client, and acquiring content data identified from live audio and video data of the anchor client;

extracting key words in the content data according to the content data and a preset semantic analysis algorithm;

obtaining a target title matched with the keyword according to the keyword;

and issuing the target topic to the anchor client.

In a third aspect, an embodiment of the present application provides a title generation method in live broadcasting, including the steps of:

responding to the content identification starting operation, and acquiring live broadcast audio and video data;

identifying content data from the live audio and video data;

generating a title generation request according to the content data, and sending the title generation request to a server;

and receiving a target topic matched by the server according to the keywords in the content data, and displaying the target topic.

In a fourth aspect, an embodiment of the present application provides a title generation system in live broadcasting, including: a host client and a server;

the anchor client is used for responding to the content identification starting operation and acquiring live broadcast audio and video data; identifying content data from the live audio and video data; generating a title generation request according to the content data, and sending the title generation request to a server;

the server is used for responding to a title generation request of the anchor client and acquiring content data identified from live audio and video data of the anchor client; extracting key words in the content data according to the content data and a preset semantic analysis algorithm; obtaining a target title matched with the keyword according to the keyword; issuing the target title to the anchor client;

and the anchor client is also used for receiving the target title matched by the server according to the keywords in the content data and displaying the target title.

In a fifth aspect, an embodiment of the present application provides a computer device, including: a processor, a memory and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method according to the first, second or third aspect when executing the computer program.

In a sixth aspect, the present application provides a computer-readable storage medium, which stores a computer program that, when executed by a processor, implements the steps of the method according to the first, second or third aspect.

In the embodiment of the application, the anchor client responds to the content identification starting operation to obtain live audio and video data, identifies the content data from the live audio and video data, generates a title generation request according to the content data and sends the title generation request to the server; the server responds to a title generation request of the anchor client, obtains content data identified from live audio and video data of the anchor client, extracts keywords in the content data according to the content data and a preset semantic analysis algorithm, obtains a target title matched with the keywords according to the keywords, and issues the target title to the anchor client; the anchor client receives the target title matched by the server according to the keywords in the content data, and displays the target title, so that the title can be quickly generated in the live broadcast process for the anchor to use, the title generation efficiency is improved, the generated title is closely related to the current live broadcast content, and the interactive experience between audiences and the anchor is effectively improved.

For a better understanding and implementation, the technical solutions of the present application are described in detail below with reference to the accompanying drawings.

Drawings

Fig. 1 is a schematic view of an application scenario of a title generation method in live broadcast according to an embodiment of the present application;

fig. 2 is a schematic flowchart of a title generation method in live broadcast according to a first embodiment of the present application;

fig. 3 is a schematic view of a live interface provided in an embodiment of the present application;

fig. 4 is a schematic flowchart of a title generation method in live broadcast according to a second embodiment of the present application;

fig. 5 is a schematic view of another application scenario of a title generation method in live broadcast according to an embodiment of the present application;

fig. 6 is a schematic flowchart of S203 in a title generation method in live broadcast according to a second embodiment of the present application;

fig. 7 is a schematic flowchart of S204 in a title generation method in live broadcast according to a second embodiment of the present application;

fig. 8 is a schematic flowchart of a title generation method in live broadcast according to a third embodiment of the present application;

fig. 9 is a schematic flowchart of a title generation method in live broadcast according to a fourth embodiment of the present application;

fig. 10 is a schematic flowchart of a title generation method in live broadcast according to a fifth embodiment of the present application;

fig. 11 is a schematic flowchart of S302 in a title generation method in live broadcast according to a fifth embodiment of the present application;

fig. 12 is another schematic flowchart of S302 in a title generation method in live broadcast according to a fifth embodiment of the present application;

fig. 13 is a timing chart of a title generation method in live broadcasting according to an embodiment of the present application;

fig. 14 is a schematic structural diagram of a title generation system in live broadcast according to a sixth embodiment of the present application;

fig. 15 is a schematic structural diagram of a computer device according to a seventh embodiment of the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

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

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

As will be appreciated by those skilled in the art, the terms "client," "terminal device," and "terminal device" as used herein include both wireless signal receiver devices, which include only wireless signal receiver devices without transmit capability, and receiving and transmitting hardware devices, which include receiving and transmitting hardware devices capable of two-way communication over a two-way communication link. Such a device may include: cellular or other communication devices such as personal computers, tablets, etc. having single or multi-line displays or cellular or other communication devices without multi-line displays; PCS (personal communications Service), which may combine voice, data processing, facsimile and/or data communications capabilities; a PDA (Personal Digital Assistant), which may include a radio frequency receiver, a pager, internet/intranet access, a web browser, a notepad, a calendar and/or a GPS (Global positioning system) receiver; a conventional laptop and/or palmtop computer or other device having and/or including a radio frequency receiver. As used herein, a "client," "terminal device" can be portable, transportable, installed in a vehicle (aeronautical, maritime, and/or land-based), or situated and/or configured to operate locally and/or in a distributed fashion at any other location(s) on earth and/or in space. The "client", "terminal Device" used herein may also be a communication terminal, a web terminal, a music/video playing terminal, such as a PDA, an MID (Mobile Internet Device) and/or a Mobile phone with music/video playing function, and may also be a smart tv, a set-top box, and the like.

The hardware referred to by the names "server", "client", "service node", etc. is essentially a computer device with the performance of a personal computer, and is a hardware device having necessary components disclosed by the von neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, an output device, etc., wherein a computer program is stored in the memory, and the central processing unit loads a program stored in an external memory into the internal memory to run, executes instructions in the program, and interacts with the input and output devices, thereby accomplishing specific functions.

It should be noted that the concept of "server" as referred to in this application can be extended to the case of a server cluster. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided, and in physical space, the servers may be independent from each other but can be called through an interface, or may be integrated into one physical computer or a set of computer clusters. Those skilled in the art will appreciate this variation and should not be so limited as to restrict the implementation of the network deployment of the present application.

Referring to fig. 1, fig. 1 is a schematic view of an application scenario of a title generation method in live broadcasting provided in an embodiment of the present application, where the application scenario includes an anchor client 101, a server 102, and a viewer client 103 provided in the embodiment of the present application, and the anchor client 101 and the viewer client 103 interact with each other through the server 102.

The anchor client 101 is a client that transmits a webcast video, and is generally a client used by an anchor (i.e., a live anchor user) in webcasting.

The viewer client 103 is a client that receives and views a webcast video, and is typically a client used by a viewer (i.e., a live viewer user) viewing a video in the webcast.

The hardware at which the anchor client 101 and the viewer client 103 are directed is essentially a computer device, and in particular, as shown in fig. 1, it may be a computer device of the type of a smart phone, smart interactive tablet, personal computer, or the like. Both the anchor client 101 and the viewer client 103 may access the internet via a known network access method to establish a data communication link with the server 102.

The server 102 is a business server, and may be responsible for further connecting related audio data servers, video streaming servers, and other servers providing related support, etc. to form a logically associated server cluster to provide services for related terminal devices, such as the anchor client 101 and the viewer client 103 shown in fig. 1.

In an optional embodiment of the present application, the server 102 may further connect to a topic library server, so that the topic library server performs topic storage, topic scoring and popularity updating, and supports topics to the server 102.

In the embodiment of the present application, the anchor client 101 and the audience client 103 may join in the same live broadcast room (i.e., a live broadcast channel), and the live broadcast room is a chat room implemented by means of an internet technology, and generally has an audio/video broadcast control function. The anchor user is live in the live room through the anchor client 101, and the audience of the audience client 103 can log in the server 102 to enter the live room to watch the live.

In the live broadcast room, interaction between the anchor user and the audience users can be realized through known online interaction modes such as voice, video, characters and the like, generally, the anchor user performs programs for the audience users in the form of audio and video streams, and economic transaction behaviors can also be generated in the interaction process. Of course, the application form of the live broadcast room is not limited to online entertainment, and can also be popularized to other related scenes, such as a knowledge live broadcast scene, a video conference scene, a product recommendation sale scene, and any other scene needing similar interaction.

The live broadcast room is in the knowledge live broadcast scene, and under the scene, the anchor client and the audience client can usually perform some question-answer interaction in the live broadcast process in the same live broadcast room so as to improve the interactive experience between the anchor and the audience.

Based on this, the embodiment of the application provides a title generation method in live broadcasting. Referring to fig. 2, fig. 2 is a schematic flowchart of a title generation method in live broadcast according to a first embodiment of the present application, where the method includes the following steps:

s101: the anchor client responds to the content identification starting operation to acquire live broadcast audio and video data; identifying content data from the live audio and video data; and generating a title generation request according to the content data, and sending the title generation request to a server.

In this embodiment, a title generation method in live broadcasting is described from two execution subjects, namely, a main broadcasting client and a server. Specifically, after the anchor clicks a live broadcast starting control in an operation interface of the anchor client, the anchor client joins the live broadcast room, and at this time, the anchor client acquires the audio and video stream and pushes the acquired audio and video stream to the server. When the audience client side joins the live broadcast room, the stream is pulled from the server, the obtained video stream is bound to the video component of the live broadcast room, and the audio stream is bound to the audio component of the live broadcast room, so that the audience can watch the real-time live broadcast in the live broadcast room in the audience client side.

In this embodiment, the mentioned live audio and video data is different from the above audio and video stream to some extent, and the live audio and video data proposed in this embodiment includes audio data and/or video data. However, the purpose of the anchor client acquiring the audio/video data is to analyze and identify the content of the audio/video data, and not to push the content to the server in real time and then send the content to the audience client for displaying and playing. Therefore, the anchor client acquires the live audio and video data only when certain preset conditions are met.

Specifically, the anchor client may capture a live broadcast picture in a video display area (i.e., a video window) of the live broadcast room based on a preset time interval, so as to obtain video data; the anchor client may also perform interception of audio data of a preset duration based on a preset time interval, so as to obtain the audio data.

The frequency of acquiring the live audio and video data provided by the embodiment by the anchor client is relatively low, so that the content identification accuracy is ensured, and meanwhile, the computing resources of the anchor client are effectively saved.

In addition, in the embodiment of the application, the anchor client responds to the content identification starting operation only after receiving the content identification starting operation, and obtains the live audio and video data.

Specifically, when the anchor user enters the live broadcast room, a content recognition switch control is displayed in a live broadcast interface of the anchor client, and the content recognition starting operation of the anchor can be received through the content recognition switch control.

Please refer to fig. 3, which is a schematic diagram of a live interface provided in an embodiment of the present application. The live interface is a graphical user interface, a video window 301 and a plurality of controls are displayed in the live interface, the controls include, but are not limited to, a content recognition switch control 302, a recording control 303, a volume adjustment control 304, a virtual gift control (not shown in fig. 3), a message input control 305, and the like, it should be noted that the display style and layout position of the video window and the controls in the live interface shown in fig. 3 are only an example, and do not have a special limiting function, and because of differences of an operating system, a software version, a channel template, and the like of a main broadcast client, the types, styles, layout positions, and the like of the video window and the controls displayed in the live interface can be changed.

In the embodiment of the application, the anchor can trigger the anchor client to acquire the live audio and video data by clicking the content identification switch control in the anchor, and perform content identification.

In other optional embodiments, since the content identification switch controls are different in category, the user may also control the content identification switch controls through a drag operation, a slide operation, and the like, thereby triggering the anchor client to acquire live audio and video data.

Specifically, from the execution perspective of the anchor client, the anchor client acquires the position of a clicked area of a user in a live interface, confirms a target control clicked by the anchor according to the position and the display position of each control in the live interface, and triggers and executes a process associated with the content recognition switch control to acquire live audio and video data in response to content recognition start operation of the anchor if the target control is the content recognition switch control.

It can be understood that after confirming that the target control clicked by the user is the content identification switch control, the anchor client further needs to determine whether the clicking operation of the anchor is a content identification opening operation or a content identification closing operation according to the state of the content identification switch control. The state of the content identification switch control may be the number of times of clicking, or whether a process associated with the content identification switch control is started.

After the live audio and video data are acquired, the anchor client identifies content data from the live audio and video data.

The live audio and video data comprises video data and/or audio data.

The video data comprises a plurality of live broadcast pictures intercepted by the anchor client in a video display area of the live broadcast room, and the audio data comprises a plurality of sections of audio intercepted by the anchor client in the live broadcast room.

For video data, the anchor client can identify content data in the video data based on a preset image identification algorithm; for audio data, the anchor client may identify content data in the video data based on a preset speech recognition algorithm. For the specific identification process, reference may be made to the explanation of step S302 in the fifth embodiment of the present application.

The content data refers to live content in a text form, the recording form of the content data in the anchor client is not limited herein, the anchor client may record the content data in a character string form, and the anchor client may record the content data in a word vector form.

In an optional embodiment, after acquiring the live audio and video data, the anchor client may send a content identification request including the live audio and video data to the server, and the server performs content identification on the live audio and video data. However, since the server needs to provide service support for a large number of clients, the content identification of the live broadcast audio/video data is performed by the anchor client, so that the server load can be reduced to a certain extent, and the overall title generation efficiency can be improved.

After identifying the content data from the live audio and video data, the anchor client generates a title generation request according to the content data, and sends the title generation request to the server.

Specifically, in this embodiment of the application, the anchor triggers the anchor client to respond to the topic generation start operation by clicking a topic generation switch control 306 shown in fig. 3 in the live interface, generates a topic generation request according to content data, and sends the topic generation request to the server.

In an alternative embodiment, the title generation request includes the content data, a host identification, and a channel identification. The anchor identification refers to a unique identification of an anchor user, and the channel identification refers to a unique identification of a live broadcast room (i.e., a network live broadcast channel). The server can confirm that the title generation request is sent by which anchor under which channel based on the anchor identification and the channel identification, so as to realize subsequent operations of scoring record, title issuing and the like.

S102: the server responds to a title generation request of the anchor client and acquires content data identified from live audio and video data of the anchor client; extracting key words in the content data according to the content data and a preset semantic analysis algorithm; obtaining a target title matched with the keyword according to the keyword; and issuing the target topic to the anchor client.

And the server acquires a title generation request sent by a main broadcasting client, and acquires the content data for the title generation request. And then, the server extracts keywords in the content data according to the content data and a preset semantic analysis algorithm.

The semantic analysis algorithm is used for performing semantic analysis on the content data and extracting key words, namely the keywords, from the content data. The keywords include at least one word.

In an optional embodiment, the semantic analysis algorithm may be stored in the server in advance, and after the server acquires the content data, the server directly calls the semantic analysis algorithm to perform semantic analysis on the content data, so as to extract the keyword.

In another alternative embodiment, the semantic analysis algorithm may be pre-stored in a data analysis server that establishes a communication link with the server, and after the server acquires the content data, the content data is sent to the data analysis server, and the keyword returned by the data analysis server is received.

The specific extraction process of the keywords is not described in detail here, and specific reference may be made to the content of step S202 in the second embodiment of the present application.

In this embodiment, after acquiring a keyword, a server acquires a target topic matched with the keyword, and issues the target topic to the anchor client.

The target topic matched with the keyword can be a target topic comprising the keyword, and can also be a target topic comprising the keyword through an associated topic tag, and each topic has at least one topic tag associated with the topic.

In an optional embodiment, before issuing the target title, the target title may be sorted according to the rating information of the target title, or the target title may be sorted according to the number of times of use of the target title, or a recommendation value of the target title is obtained according to the number of times of use of the target title and the rating information, the target title is sorted according to the recommendation value, and then the server issues the sorted target title to the anchor client according to the channel identifier and the anchor identifier.

The scoring information refers to the scoring of each target topic by the anchor collected by the anchor client after the target topic is selected for question-answer interaction with the audience. The scoring information of each topic can be accumulated and stored in the server, and when any anchor scores the target topic, the server updates the scoring information of the target topic.

The number of usage times is the number of times the server accumulates to use the topic. Specifically, after any anchor selects a target topic to perform question-answer interaction with audiences, the number of times of use of the topic is increased by one.

The recommendation degree value is a numerical value comprehensively calculated according to the grading information and the use times, and the target titles are sequenced by using the recommendation degree value, so that the titles which are closely related to the live broadcast content and are interested by the audience can be more easily acquired by the anchor.

S103: and the anchor client receives a target title matched by the server according to the keywords in the content data and displays the target title.

In an alternative embodiment, the anchor client, upon receiving the target title, can display the target title in a video window within the live interface. Specifically, the anchor client may mix the target topic with the video data stream, and load the target topic as a layer on a video data frame, so that the target topic can be watched by the audience in real time.

In another optional embodiment, after receiving the target title, the anchor client binds the target title with the target title display control, and the anchor client responds to the target title display operation and displays a target title list containing the target title in a live interface of the anchor client. At this time, only the anchor can be viewed in the live broadcast room, and the selected title can be sent to the audience client in the live broadcast room through the server after being screened by the anchor.

Referring to fig. 4, fig. 4 is a flowchart illustrating a title generation method in live broadcasting according to a second embodiment of the present application, where the method is mainly executed by the server 102 in fig. 1, and as shown in fig. 4, the method includes the following steps:

s201: and responding to a title generation request of a main broadcast client, and acquiring content data identified from the live audio and video data of the main broadcast client.

S202: and extracting the keywords in the content data according to the content data and a preset semantic analysis algorithm.

S203: and acquiring a target title matched with the keyword according to the keyword.

S204: and issuing the target topic to the anchor client.

The specific execution process of step S201 is already described in step 102 of the first embodiment, and is not described herein again.

The following describes the keyword extraction process in step S202 in detail, specifically as follows:

in an optional embodiment, the server may input the content data into a pre-trained semantic analysis model, and obtain a criticality value of each word in the content data output by the semantic analysis model. The semantic analysis model is a pre-trained model based on text data in a preset text library. The semantic analysis model may be any natural language processing model, such as: the long and short term memory network model, the recurrent neural network model, etc. are not limited in detail herein.

And labeling corresponding keywords on each text data in the text library. And iteratively training parameters of the semantic analysis model through each text data in the text base and the labeled keywords until preset semantic analysis model training convergence conditions are met, and obtaining pre-trained parameters of the semantic analysis model.

Specifically, the training process of the semantic analysis model may be performed in a server, or may be performed in a training device that establishes a communication link with the server, and then the pre-trained semantic analysis model parameters are transplanted to the server.

The semantic analysis model outputs the probability that each word in the content data is a keyword, in this embodiment, the probability is used as a keyword value, and then, the word with the keyword value larger than a preset keyword threshold value in each word of the server is used as the keyword.

In another optional embodiment, the server may also perform word segmentation on the content data by using an existing word segmentation tool, perform part-of-speech tagging on the word after word segmentation, calculate a keyword value of each word in the content data according to the part-of-speech of the word and the occurrence frequency of the word in the content data, and finally extract a keyword of which the keyword value meets a preset keyword threshold value from the word.

In this embodiment, based on a preset semantic analysis algorithm, the accuracy of the obtained keywords can be improved, so that target questions which are relatively closely related to the live broadcast content are matched, and the effect of question answering interaction is favorably improved.

In order to further exert the subjective initiative of the anchor, and obtain a target title that is more matched with the current live broadcast situation and better conforms to the expectation of the anchor, after the server executes step S202, the following operations may also be executed:

the method comprises the steps that a server receives a topic generation mode instruction sent by a main broadcasting client, then a corresponding topic generation mode is obtained according to the topic generation mode instruction, when the topic generation mode is a non-full-automatic mode, keywords are sent to the main broadcasting client, and selected keywords and user-defined keywords returned by the main broadcasting client are received.

In the embodiment of the application, the title generation mode instruction comprises a channel identifier, a main broadcasting identifier and a title generation mode, and the title generation mode comprises a full-automatic mode and a non-full-automatic mode.

And in the full-automatic mode, after extracting the keywords in the content data, the server directly searches the target titles matched with the keywords.

And after extracting the keywords in the content data, the server sends the keywords to the anchor client in a non-full-automatic mode, so that the anchor client displays a keyword selection control and a user-defined keyword input control containing the keywords in a live interface, and receives the selected keywords and the user-defined keywords returned by the anchor client.

In this embodiment, if the topic generation mode included in the topic generation mode instruction sent by the anchor client is a non-fully automatic mode, the anchor may further filter and add the keywords of the content data extracted by the server through the anchor client, so as to obtain a target topic that is more matched with the current live broadcast situation and more conforms to the expectation of the anchor.

The following describes in detail the process of matching the target title in step S203, specifically as follows:

in the first embodiment, in order to reduce the load of the server 102, please refer to fig. 5, where fig. 5 is a schematic view of another application scenario of the topic generation method in live broadcasting provided in the embodiment of the present application, in which the server 102 is further connected to the topic library server 104 to form a logically associated service cluster, so that the topic library server 104 performs topic storage, topic scoring and heat update, and performs topic support on the server 102.

Specifically, in an alternative embodiment, referring to fig. 6, step S203 includes steps S2031 to S2032, as follows:

s2031: and sending the keywords to a question bank server, so that the question bank server searches whether a target question matched with the keywords exists in a preset question bank, if so, acquiring the question matched with the keywords and called from the preset question bank by the question bank server, and if not, acquiring the question matched with the keywords and searched from a network by the question bank server.

In the embodiment of the application, the question bank is pre-established in the question bank server, and a plurality of questions, scoring information of each question and the use times of each question are stored in the question bank. The question bank server can dynamically update the questions in the question bank, the grading information of each question and the using times of each question according to the user data generated in the live broadcast process.

After receiving the keywords, the question bank server searches whether a target question matched with the keywords exists in the question bank. In an alternative embodiment, the target topic that matches the keyword can refer to a target topic that includes the keyword. In another alternative embodiment, several topics stored in a topic each have at least one topic tag associated therewith. Thus, a target topic that matches the keyword can also refer to a target topic for which the associated topic tag includes the keyword.

And if the target questions matched with the keywords exist in a preset question bank, the question bank server calls the questions matched with the keywords from the preset question bank and sends the questions matched with the keywords to the server. And if the preset question bank does not have the target question matched with the keyword, the question bank server searches the question matched with the keyword from the network and sends the question matched with the keyword to the server.

In some cases, the question bank server has a question which cannot be searched from the network and is matched with the keyword, and at the moment, the question bank server stores the keyword, and sends the keyword to the anchor client through the server or the operator client through the server after the live broadcast is finished, so that the anchor or the operator can issue the question according to the keyword and update the question to the question bank server.

In an optional embodiment, after the question bank server obtains the target topics matching the keywords, the target topics are further sorted.

Specifically, the question bank server obtains the target questions after screening and sorting according to the target questions and a preset initial screening algorithm.

The initial screening and sorting algorithm may sort the target titles according to the rating information of the target titles or the number of times of using the target titles, or may obtain a recommendation value of the target titles according to the rating information of the target titles and the number of times of using the target titles, and sort the target titles based on the recommendation value of the target titles.

And the scoring information is the scoring of each target topic by the anchor collected by the anchor client after the anchor selects the target topic to perform question-answer interaction with the audience. The scoring information of each topic can be accumulated and stored in the topic library server, and after any anchor scores a target topic, the topic library server updates the scoring information of the target topic.

The using times are the accumulated using times of the title by the title library server. Specifically, after any anchor selects a target topic to perform question-answer interaction with audiences, the number of times of use of the target topic is increased by one.

S2032: and receiving the target topic which is returned by the topic library server and matched with the keyword.

The server receives the target title matched with the keyword returned by the title library server.

It can be understood that the server will send the channel identifier and the anchor identifier when sending the keyword to the question bank server, and then will also send the channel identifier and the anchor identifier when the question bank server returns the target question matching the keyword. So that the server can confirm to which anchor client the target title is to be sent for display.

In this embodiment, a service cluster logically associated with each other is formed by establishing a communication link between the server 102 and the question bank server 104, so that keyword matching and updating of the data related to the question are performed by the question bank server, the load of the server is reduced to a certain extent, the efficiency of generating the question is further improved, and interference on real-time live broadcast is reduced. In addition, the target topics are sorted and screened before the topic library server returns the target topics, so that the target topics can better meet the requirements of the anchor, are close to the current live broadcast content and the hot topics, and the interactivity with audiences is improved.

The following describes in detail the process of issuing the target topic to the anchor client in step S204, specifically as follows:

in an optional embodiment, before issuing the target topic to the anchor client, the server may further filter the target topic to remove the topic unrelated to the live topic, thereby reducing the filtering operation of the anchor in the anchor client and improving the intelligence of topic generation. Referring to fig. 7, step S204 includes steps S2041 to S2042, which are as follows:

s2041: and acquiring a live broadcast theme of a live broadcast room created by the anchor.

In webcasting, some live activities with live topics are often performed, such as: intellectual property themes, relatives themes, or mental health themes, etc.

Therefore, because the live broadcast topics of the live broadcast rooms are different, when the anchor broadcasts in the live broadcast room, the required topics are more targeted, and therefore, before the target topics are issued to the anchor client, the server needs to acquire the live broadcast topics of the live broadcast room created by the anchor.

S2042: and screening and sequencing the target titles according to a preset screening and sequencing algorithm related to the live broadcast topics, and issuing the screened and sequenced target titles to the anchor client.

In this embodiment, each live topic is a corresponding screening and sorting algorithm, and a target topic more suitable for the live topic can be screened from the target topics through the screening and sorting algorithm related to the live topic.

Specifically, the screening and sorting algorithm related to the live broadcast theme may be to search for a target title including more target vocabularies according to preset target vocabularies related to the live broadcast theme and the target title, set a higher display priority of a target body including more target vocabularies, remove a target title not including target vocabularies, sort the target titles according to the display priority, obtain the screened and sorted target titles, and send the screened and sorted target titles to the anchor client.

In the embodiment, the target titles can be further screened under different live broadcast themes to remove titles irrelevant to the live broadcast themes, so that the screening operation of the anchor in the anchor client is further reduced, and the intelligence and efficiency of title generation are improved.

Referring to fig. 8, fig. 8 is a schematic flow chart of a title generation method in live broadcasting according to a third embodiment of the present application, where the method is mainly executed by the server 102 in fig. 1, and is different from the second embodiment in that steps S201 to S204 are followed by steps S205 to S208, which are as follows:

s205: and receiving the selection information of the target topic returned by the anchor client.

In this embodiment, a server issues a target title to a anchor client, and after receiving the target title, the anchor client generates a title selection control according to the target title and displays the title selection control in a live interface of the anchor client. And the title selection control comprises options corresponding to the target title.

And the anchor client receives the trigger operation of the anchor through the title selection control, and if the option of the target title is triggered, the target title is confirmed to be the selected title, so that the selection information of the target title is obtained.

The selection information of the target topic comprises an identifier of the selected topic. It will be appreciated that the identification is a unique identification of the selected topic.

And then, the anchor client sends the selection information of the target topic to the server, and the server receives the selection information of the target topic returned by the anchor client.

S206: an identification of a live room created by a anchor is obtained.

In this embodiment, the identifier of the live broadcast room is a channel identifier. When the client sends the selection information of the target title to the server, the channel identifier and the anchor identifier are also sent together.

S207: and acquiring the audience client in the live broadcast room according to the identification of the live broadcast room.

S208: and issuing the selected title to the audience client in the live broadcast room according to the selection information of the target title.

And the server issues the corresponding selected title to the audience client in the live broadcast room according to the identification of the selected title in the selection information, so that the audience client displays the selected title in a live broadcast interface.

In an alternative embodiment, the manner of displaying the selected topic by the viewer client may be to display a popup window containing the selected topic, or to mix the selected topic with the video data stream, and load the selected topic as a layer onto a video data frame, so that the viewer can view the selected topic in a live interface of the viewer client through any manner.

Referring to fig. 9, fig. 9 is a flowchart illustrating a title generation method in live broadcasting according to a fourth embodiment of the present application, where the method is mainly executed by the server 102 in fig. 1, and the difference from the third embodiment is that after the server executes steps S205 to S208, the server needs to execute steps S209 to S212, where steps S209 to S212 are specifically as follows:

s209: and receiving the grading information of the selected topic returned by the anchor client.

After answering interaction or live broadcasting is finished, a anchor client displays a topic grading control, receives grading information of each selected topic by the anchor through the topic grading control, sends the grading information of the selected topic to a server, and the server receives the grading information of the selected topic returned by the anchor client.

S210: and updating the use times and the grading information of the selected topic according to the selected topic and the grading information of the selected topic.

In an optional embodiment, the server updates the use times and the scoring information of the selected topic according to the selected topic and the scoring information of the selected topic.

Specifically, the server may add the current rating information of the selected topic to the original rating information of the selected topic, and add one to the number of times of use of the selected topic, thereby updating the rating information and the number of times of use of the selected topic.

In another optional embodiment, the server generates a first update instruction according to the selected topic and the score information of the selected topic, and sends the first update instruction to the question bank server, so that the question bank server updates the number of times of use and the score information of the selected topic, and a specific update process of the question bank server is consistent with the above, which is not described herein again.

S211: and updating the recommendation value of the selected topic according to the using times and the grading information.

If the target questions are sorted by the recommendation values, the server further needs to update the recommendation values of the selected questions according to the use times and the grading information, or the server generates a second updating instruction according to the use times and the grading information, sends the second updating instruction to the question bank server, enables the question bank server to calculate the recommendation values of the selected questions according to the second updating instruction, and accumulates the recommendation values of the selected questions to the original recommendation values of the selected questions.

S212: and when the next response is made to the title generation request of the anchor client, sequencing the target titles according to the updated recommendation degree value.

In this embodiment, the server can receive feedback information of the anchor on the selected topic, and update the rating information, the use times and the recommendation value of the selected topic, so as to provide better information support when the sorted topics are screened next time, and provide better topics for the anchor for question-answer interaction.

Referring to fig. 10, fig. 10 is a schematic flowchart of a title generation method in live broadcast according to a fifth embodiment of the present application, where the method is mainly executed by the anchor client 101 in fig. 1, and the method includes steps S301 to S304, which are as follows:

s301: and responding to the content identification starting operation to acquire the live audio and video data.

S302: and identifying content data from the live audio and video data.

S303: and generating a title generation request according to the content data, and sending the title generation request to a server.

S304: and receiving a target topic matched by the server according to the keywords in the content data, and displaying the target topic.

The specific operation of responding to the content identification start operation in step S301 to acquire live audio/video data is described in detail in step S101 of the first embodiment, and is not described herein again.

In the following, how to identify the content data from the live audio/video data in step S302 is described as follows:

referring to fig. 3, the live interface shown in fig. 3 is substantially a live interface corresponding to the education template, and the live interface further includes a lesson taking control 307, a release switch control (i.e., the tutorial courseware control 306 shown in fig. 3), and a file uploading control (i.e., the local uploading control 309 shown in fig. 3). Different from a conventional live broadcast interface, under the condition that a channel template is an education template, live broadcast can be started only after an anchor clicks a class-taking control in a live broadcast interface instead of immediately starting the live broadcast after the anchor clicks a live broadcast control to enter a live broadcast room, and the purpose of the setting is to reserve a certain time for the anchor to upload files and other relevant preparations.

Specifically, after the anchor client side joins the live broadcast room, a file uploading control and a release switch control are displayed in a live broadcast interface of the anchor client side, the trigger operation of a user is received through the release switch control, the current live broadcast mode is set to be the release mode, the file uploading control receives the uploading operation, a file to be released is obtained, and the content of the file to be released is displayed in a video display area of the live broadcast interface.

The file uploading control comprises a local uploading control and a network disk uploading control. The anchor can upload files in different ways on the live interface.

In addition, the live interface further includes a camera opening control (i.e., the camera control 310 shown in fig. 3), the camera opening control receives a trigger operation of a user, the current live mode is set to be a non-delivery mode, the camera is started, and a video stream shot by the camera is displayed in a video display area of the live interface.

When the live broadcast mode is in a release mode or a non-release mode, the anchor client identifies content data from the live broadcast audio/video data in different modes, specifically as follows:

referring to fig. 11, step S302 includes steps S3021 to S3022 as follows:

s3021: if the current live broadcast mode is an input mode, acquiring a first live broadcast picture in the video display area; and the first direct playing picture is the content of the file to be launched.

And the anchor client judges whether the current live broadcast mode is the release mode, and if the current live broadcast mode is the release mode, the position of the video display area in the live broadcast interface, the length of the video display area and the width of the video display area are obtained.

In an optional embodiment, the position of the video display area in the live interface comprises a relative distance between the upper left corner of the video display area and a frame on the live interface and a relative distance between the upper left corner of the video display area and the left frame of the live interface.

And the anchor client side intercepts a first direct playing picture from the video display area according to the content data identified from the live audio and video data, and at the moment, the first direct playing picture is the content of the file to be launched.

S3022: and identifying content data in the first direct broadcasting picture according to the first direct broadcasting picture and a preset image content identification algorithm.

The preset image content Recognition algorithm may be any Optical Character Recognition algorithm (OCR), and is not limited herein.

The image content identification algorithm may be preset in the anchor client, and the anchor client calls the image content identification algorithm after acquiring the first live image, so as to obtain the content data in the first live image.

Obviously, the content data also refers to a text-form live content in this embodiment, a recording form of the content data in the anchor client is not limited herein, the anchor client may record the content data in a character string form, or the anchor client may record the content data in a word vector form.

In this embodiment, the anchor client determines the current live mode first, and then selects the identification mode corresponding to the current live mode to obtain the content data, thereby improving the accuracy of the content data.

Referring to fig. 12, step S302 includes steps S3023 to S3025 as follows:

s3023: if the current live broadcast mode is a non-delivery mode, acquiring a second live broadcast picture in the video display area; and the second live broadcast picture is a live broadcast camera shooting picture.

And the anchor client judges whether the current live mode is the release mode, and if the current live mode is the non-release mode, the position of the video display area in the live interface, the length of the video display area and the width of the video display area are obtained.

And the anchor client side intercepts a second live broadcast picture from the video display area according to the content data identified from the live broadcast audio and video data, and at the moment, the second live broadcast picture is a live broadcast camera shooting picture.

S3024: and segmenting a character area in the second live broadcast picture.

In the non-delivery mode, the second picture is a picture shot by the live-broadcast camera, so that the content of the second live-broadcast picture cannot be directly identified in the non-delivery mode, but the second live-broadcast picture needs to be firstly subjected to picture segmentation, positioned to a character area in the second live-broadcast picture and then subjected to image identification.

For example: in the non-release mode, the anchor may use the blackboard for live broadcast, and then the second live broadcast picture obtained includes a non-blackboard area and a blackboard area, and the blackboard area includes a non-text area and a text area. If the second live broadcast picture is directly live broadcast, the picture of the non-character area will affect the accuracy of image recognition.

In this embodiment, the text area may be used as a foreground area for image segmentation, and the non-text area may be used as a background area for image segmentation. The anchor client can divide a foreground area, namely a character area, from the second live broadcast picture according to the second live broadcast picture and a preset image division model.

In an alternative embodiment, the image segmentation model is a pre-trained image segmentation model, the training data of the image segmentation model may be images labeled with a foreground region and a background region, and the training method may be performed by using a supervised learning method, which is not limited in detail herein.

S3025: and identifying content data in the second live broadcast picture according to a character area in the second live broadcast picture and a preset image content identification algorithm.

The content data is identified from the text area in the same manner as the content data is identified from the first live view, and the details are not repeated.

In the process that the anchor client executes step S303 to generate a title generation request according to the content data and sends the title generation request to the server, the anchor client can also select a title generation mode.

Specifically, the method comprises the following steps: displaying a title generation mode selection control in a live interface of a main broadcasting client; the title generation mode selection control comprises a full-automatic mode option and a non-full-automatic mode option. And then, receiving trigger operation of a main broadcast through the title generation mode selection control, and sending a title mode generation instruction containing a title generation mode to a server.

In an optional embodiment, the topic generation mode selection control can be used as a next-level control of the topic generation switch control, that is, after the topic generation switch control in the live interface is clicked, the next-level control in the topic generation switch control is displayed as the topic generation mode selection control.

If the title mode generation instruction sent to the server contains a non-full-automatic title generation mode, receiving the keyword sent by the server by the anchor client, and displaying a keyword selection control and a user-defined keyword input control in a live interface of the anchor client, wherein the keyword selection control comprises an option corresponding to the keyword received from the server; then receiving trigger operation of an anchor through the keyword selection control, determining the keyword as a selected keyword if an option corresponding to the keyword is triggered, and acquiring an input custom keyword if the custom keyword input control receives input operation; and finally, sending the selected keywords and the user-defined keywords to the server.

In this embodiment, the anchor can automatically select the topic generation mode, and when the anchor selects the non-full-automatic mode, the anchor can further filter and add the keywords of the content data extracted by the server through the anchor client, so that the target topics which are more matched with the current live broadcast condition and more accord with the expectation of the anchor can be obtained.

The specific implementation process of receiving the target title matched by the server according to the keyword in the content data in step S304 and displaying the target title has been described in detail in step S103 of the first embodiment, and is not described herein again.

In an alternative embodiment, after the anchor client executes 304, the method further includes the steps of: generating a target title selection control according to a target title sent by a server; displaying the target title selection control in a live interface of a main broadcasting client; the target topic selection control comprises options corresponding to all the target topics; then, receiving a trigger operation of a main broadcasting through the target title selection control, and if the option of the target title is triggered, determining the target title as a selected title to obtain selection information of the target title; finally, the selection information of the target title is sent to the server, so that the server issues the selected title to a viewer client in a live broadcast room; wherein the live broadcast room is a live broadcast room created by the anchor.

By displaying the target title selection control in the anchor client, the anchor further selects the target title, so that the selected title issued to the audience client can generate better interaction effect, and the interaction enthusiasm between the audience and the anchor is improved.

Please refer to fig. 13, which is a timing diagram illustrating a title generation method in live broadcasting according to an embodiment of the present application. Referring to the timing diagram, the overall flow of the title generation method in live broadcast is more intuitively explained so as to facilitate understanding of the technical scheme provided by the application, in the embodiment of the application, a main broadcast client and a spectator client are added into the same live broadcast room, the main broadcast client acquires audio and video streams in real time, packages the audio and video streams and pushes the audio and video streams to a server, and the spectator client pulls the audio and video streams from the server and analyzes and displays the audio and video streams; the anchor client side responds to the content identification starting operation, acquires live audio and video data, identifies content from the live audio and video data, generates a title generation request according to the content data, and sends the title generation request containing the content data to the server; the server responds to a title generation request of the anchor client and extracts keywords in the content data according to the content data and a preset semantic analysis algorithm; the server receives a topic generation mode instruction sent by the anchor client, then obtains a corresponding topic generation mode according to the topic generation mode instruction, and sends a keyword to the anchor client when the topic generation mode is a non-full-automatic mode; the anchor client receives the keywords, displays a keyword selection control and a user-defined keyword input control containing the keywords, acquires the selected keywords and the user-defined keywords through the keyword selection control and the user-defined keyword input control, and sends the selected keywords and the user-defined keywords to the server; the server forwards the selected keywords and the user-defined keywords to the question bank server; the question bank server can search a target question matched with the selected keyword and the user-defined keyword in the question bank, screen and sort the target question according to the grading information, the using times or the recommendation degree of the target question, and send the screened and sorted target question to the server; the server receives the screened and sequenced target titles and sends the screened and sequenced target titles to the anchor client; the anchor client receives the screened and sequenced target titles, generates a target title selection control according to the target titles, displays the target title selection control in a live broadcast interface of the anchor client, acquires selection information of the target titles through the target title selection control, sends the selected information of the target titles to a server, and the server issues the selected titles to audience clients added in the anchor room according to the selected information of the target titles; after answering interaction or live broadcasting is finished, the anchor client displays a question grading control, receives grading information of each selected question by the anchor through the question grading control, and sends the grading information of the selected question to a server; the server receives the grading information of the selected topic returned by the anchor client and forwards the grading information to the topic library server; and the question bank server updates the use times, the grading information and the recommendation value of the selected question according to the grading information of the selected question so as to be used when the next question generation service is provided for the anchor.

The method provided by the embodiment of the application can quickly generate the title for the anchor in the live broadcasting process, so that the title generation efficiency is improved, the generated title is closely related to the content of the current live broadcasting, the interactive experience between audiences and the anchor is effectively improved, and in addition, a huge knowledge system can be formed through the accumulation of anchor preference data, so that the validity of the title generation method is further improved, the title is more suitable for the current live broadcasting scene, the requirements of the anchor and the audiences are better met, and a more positive interactive effect is generated.

It should be noted that some optional implementations in the first to fifth embodiments are not individually shown in the timing chart, but all main steps in the title generation method in live broadcast are already shown in fig. 13, fig. 13 is only used to help understanding the technical solution of the present application, and implementations not shown in the figures are still within the protection scope of the present application.

Please refer to fig. 14, which is a schematic structural diagram of a title generation system in live broadcasting according to a sixth embodiment of the present application. As shown in fig. 14, the title generation system 14 in live broadcast includes: an anchor client 141 and a server 142.

The anchor client 141 is configured to respond to a content identification start operation and acquire live audio/video data; identifying content data from the live audio and video data; and generating a title generation request according to the content data, and sending the title generation request to the server 142.

The server 142 is configured to respond to a title generation request of the anchor client 141, and acquire content data identified from live audio/video data of the anchor client 141; extracting key words in the content data according to the content data and a preset semantic analysis algorithm; obtaining a target title matched with the keyword according to the keyword; and issuing the target topic to the anchor client 141.

The anchor client 141 is further configured to receive a target topic matched by the server 142 according to the keyword in the content data, and display the target topic.

Fig. 15 is a schematic structural diagram of a computer device according to a seventh embodiment of the present application. As shown in fig. 15, the computer device 15 may include: a processor 150, a memory 151, and a computer program 152 stored in the memory 151 and operable on the processor 150, for example: a title generation method in live broadcasting; the steps of the first embodiment are implemented when the processor 150 executes the computer program 152; alternatively, the processor 150 implements the steps of the second to fourth embodiments when executing the computer program 152, such as the steps S201 to S204 shown in fig. 4; alternatively, the processor 150 implements the steps in the fifth embodiment, such as the steps S301 to S304 shown in fig. 10, when executing the computer program 152;

the processor 150 may include one or more processing cores, among others. The processor 150 is connected to various parts in the computer device 15 by various interfaces and lines, executes various functions of the computer device 15 and processes data by executing or executing instructions, programs, code sets or instruction sets stored in the memory 151 and calling data in the memory 151, and optionally, the processor 150 may be implemented in at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), Programmable Logic Array (PLA). The processor 150 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing contents required to be displayed by the touch display screen; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 150, but may be implemented by a single chip.

The Memory 151 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 151 includes a non-transitory computer-readable medium. The memory 151 may be used to store instructions, programs, code sets or instruction sets. The memory 151 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions, etc.), instructions for implementing the various method embodiments described above, and the like; the storage data area may store data and the like referred to in the above respective method embodiments. The memory 151 may optionally be at least one memory device located remotely from the processor 150.

The embodiment of the present application further provides a computer storage medium, where the computer storage medium may store a plurality of instructions, where the instructions are suitable for being loaded by a processor and executing the method steps in the foregoing embodiment, and a specific execution process may refer to specific descriptions of the foregoing embodiment, which is not described herein again.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

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

The present application is not limited to the above-described embodiments, and various changes and modifications to the present application are intended to be included within the scope of the claims and the equivalent technology of the present application if they do not depart from the spirit and scope of the present application.

Claims

1. A title generation method in live broadcasting is characterized by comprising the following steps:

2. A title generation method in live broadcasting is characterized by comprising the following steps:

obtaining a target title matched with the keyword according to the keyword;

and issuing the target topic to the anchor client.

3. The title generation method in live broadcasting according to claim 2, wherein before obtaining a target title matching the keyword according to the keyword, the method comprises:

receiving a title generation mode instruction sent by the anchor client;

acquiring a corresponding topic generation mode according to the topic generation mode instruction, and sending the keyword to the anchor client when the topic generation mode is a non-full-automatic mode;

and receiving the selected keywords and the self-defined keywords returned by the anchor client.

4. The title generation method in live broadcasting of claim 2, wherein the step of obtaining the target title matched with the keyword according to the keyword comprises the steps of:

sending the keyword to a question bank server, enabling the question bank server to search whether a target question matched with the keyword exists in a preset question bank, if so, acquiring a question matched with the keyword and called from the preset question bank by the question bank server, and if not, acquiring a question matched with the keyword and searched from a network by the question bank server;

and receiving the target topic which is returned by the topic library server and matched with the keyword.

5. The title generation method in live broadcasting according to claim 4, characterized in that: the target titles are obtained by screening the question bank server according to a preset initial screening and sorting algorithm, wherein the initial screening and sorting algorithm is used for sorting the target titles according to the grading information of the target titles or the using times of the target titles, or obtaining the recommendation value of the target titles according to the grading information of the target titles and the using times of the target titles, and sorting the target titles based on the recommendation value of the target titles.

6. The title generation method in live broadcasting of claim 2, wherein said issuing said target title to a main broadcasting client comprises the steps of:

acquiring a live broadcast theme of a live broadcast room created by a main broadcast;

and screening and sequencing the target titles according to a preset screening and sequencing algorithm related to the live broadcast topics, and issuing the screened and sequenced target titles to the anchor client.

7. The title generation method in live broadcasting according to any one of claims 2 to 6, wherein after the target title is issued to a main broadcasting client, the method comprises the steps of:

receiving selection information of the target title returned by the anchor client;

acquiring an identifier of a live broadcast room created by a main broadcast;

acquiring audience client sides in the live broadcast room according to the identification of the live broadcast room;

and issuing the selected title to the audience client in the live broadcast room according to the selection information of the target title.

8. The title generation method in live broadcasting according to claim 7, further comprising the steps of:

receiving scoring information of the selected topic returned by the anchor client;

and updating the use times and the grading information of the selected topic according to the selected topic and the grading information of the selected topic.

9. The title generation method in live broadcasting according to claim 8, further comprising, after updating the number of times of use of the selected title and the rating information according to the selected title and the rating information of the selected title, the steps of:

updating the recommendation value of the selected topic according to the using times and the grading information;

and when the next response is made to the title generation request of the anchor client, sequencing the target titles according to the updated recommendation degree value.

10. The title generation method in live broadcasting according to claim 2, wherein the extracting keywords from the content data according to the content data and a preset semantic analysis algorithm comprises:

acquiring words in the content data and a criticality value of each word in the content data according to the content data and a preset semantic analysis algorithm;

and extracting the keywords of which the keyword values meet a preset keyword threshold value from the words.

11. A title generation method in live broadcasting is characterized by comprising the following steps:

identifying content data from the live audio and video data;

12. The title generation method in live broadcasting according to claim 11, wherein said acquiring live broadcasting audio/video data in response to a content recognition start operation comprises the steps of:

displaying a content identification switch control in a live interface of the anchor client, receiving content identification starting operation of the anchor through the content identification switch control, and acquiring the live audio and video data.

13. The title generation method in live broadcasting according to claim 11, wherein the step of generating a title generation request from the content data further comprises:

displaying a title generation mode selection control in a live interface of the anchor client; the title generation mode selection control comprises a full-automatic mode option and a non-full-automatic mode option;

and receiving trigger operation of a main broadcast through the title generation mode selection control, and sending a title mode generation instruction containing a title generation mode to a server.

14. The title generation method in live broadcasting of claim 13, wherein if a title mode generation instruction sent to a server includes a non-full-automatic title generation mode, the keyword sent by the server is received;

displaying the keyword selection control and a user-defined keyword input control in a live interface of the anchor client, wherein the keyword selection control comprises options corresponding to the keywords received from a server;

receiving trigger operation of a main broadcast through the keyword selection control;

if the option corresponding to the keyword is triggered, determining the keyword as the selected keyword;

if the user-defined keyword input control receives an input operation, acquiring an input user-defined keyword;

and sending the selected keywords and the custom keywords to the server.

15. A method of title generation in a live broadcast according to any of claims 11 to 14, wherein before identifying content data from the live audiovisual data, comprising the steps of:

displaying a release switch control and a file uploading control in a live broadcast interface of the anchor client;

receiving a triggering operation of a user through the release switch control, and setting a current live broadcast mode as a release mode;

and receiving an uploading operation through the file uploading control, acquiring a file to be launched, and displaying the content of the file to be launched to a video display area of the live broadcast interface.

16. A method of generating titles in a live broadcast according to claim 15, wherein said identifying content data from the live audio/video data comprises the steps of:

if the current live broadcast mode is an input mode, acquiring a first live broadcast picture in the video display area; the first direct playing picture is the content of the file to be launched;

and identifying content data in the first direct broadcasting picture according to the first direct broadcasting picture and a preset image content identification algorithm.

17. A method of generating titles in a live broadcast according to claim 15, wherein said identifying content data from the live audio/video data comprises the steps of:

if the current live broadcast mode is a non-delivery mode, acquiring a second live broadcast picture in the video display area; the second live broadcast picture is a shot picture of a live broadcast camera;

segmenting a character area in the second live broadcast picture;

and identifying content data in the second live broadcast picture according to a character area in the second live broadcast picture and a preset image content identification algorithm.

18. A method for generating titles in a live broadcast according to any one of claims 11 to 14, wherein after receiving target titles matched by the server according to the keywords in the content data, the method comprises the steps of:

generating a target question selection control according to the target question; the target topic selection control comprises options corresponding to all the target topics;

displaying the target title selection control in a live interface of the anchor client;

receiving a trigger operation of a main broadcasting through the target title selection control, and if the option of the target title is triggered, determining the target title as a selected title to obtain selection information of the target title;

sending the selection information of the target title to the server, so that the server sends the selected title to a viewer client in a live broadcast room; wherein the live broadcast room is a live broadcast room created by the anchor.

19. A title generation system in live broadcasting, comprising: a host client and a server;

20. A computer device, comprising: processor, memory and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to claim 1, any of claims 2 to 10 or any of claims 11 to 18 when executing the computer program.

21. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to claim 1, any one of claims 2 to 10 or any one of claims 11 to 18.