CN117714804A - Live interaction method and device, computer equipment and storage medium - Google Patents
Live interaction method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN117714804A CN117714804A CN202311750380.6A CN202311750380A CN117714804A CN 117714804 A CN117714804 A CN 117714804A CN 202311750380 A CN202311750380 A CN 202311750380A CN 117714804 A CN117714804 A CN 117714804A
- Authority
- CN
- China
- Prior art keywords
- interaction
- content
- interactive
- event
- queue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003993 interaction Effects 0.000 title claims abstract description 325
- 238000000034 method Methods 0.000 title claims abstract description 57
- 230000007123 defense Effects 0.000 claims abstract description 66
- 230000001960 triggered effect Effects 0.000 claims abstract description 14
- 230000002452 interceptive effect Effects 0.000 claims description 256
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 abstract description 13
- 230000008569 process Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 10
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004883 computer application Methods 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 241000220317 Rosa Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
The application discloses a live broadcast interaction method, a live broadcast interaction device, computer equipment and a storage medium, and relates to the technical field of video processing. The method comprises the following steps: acquiring an interaction event triggered by a viewer user in a live broadcast room, wherein the interaction event carries interaction content and user identification information of the viewer user; judging whether the interaction event is an effective interaction event by adopting a preset defense distribution strategy; and if the interaction event is an effective interaction event, replying the interaction content in the interaction event through a large language model to obtain replied content. The method and the device can improve interactivity between the anchor and the audience user.
Description
Technical Field
The present disclosure relates to the field of video processing technologies, and in particular, to a live interaction method, a live interaction device, a computer device, and a storage medium.
Background
Live video application has become an indispensable part of life of people, and live broadcasting changes life style of people through entertainment, sharing, shopping and the like of live broadcasting people.
Live broadcasting is carried out by a live broadcasting AI virtual person in a new live broadcasting mode, in order to improve interactivity between the virtual person and audience users in the live broadcasting process, simple reply content can be set for the virtual person in advance, trigger keywords are set in advance, and under the condition that bullet screen information sent by the audience users is the trigger keywords, the virtual person can send out voice based on the reply content corresponding to the trigger keywords to realize interactivity with the audience.
However, the inventor finds that in the live broadcast process of the existing virtual person, the virtual person cannot conduct a dialogue with the audience user except the trigger keywords, the interactivity between the virtual person and the audience user is low, and the user experience is poor.
Disclosure of Invention
In view of the foregoing, a live interaction method, apparatus, computer device and storage medium are now provided to solve the above-mentioned problems.
The application provides a live interaction method, which comprises the following steps:
acquiring an interaction event triggered by a viewer user in a live broadcast room, wherein the interaction event carries interaction content and user identification information of the viewer user;
judging whether the interaction event is an effective interaction event by adopting a preset defense distribution strategy;
and if the interaction event is an effective interaction event, replying the interaction content in the interaction event through a large language model to obtain replied content.
Optionally, the method further comprises:
and discarding the interaction event if the interaction event is an invalid interaction event.
Optionally, the arming policy includes at least one arming policy, and the determining, with the preset arming policy, whether the interaction event is a valid interaction event includes:
And judging whether the interaction event is an effective interaction event by adopting the at least one defense distribution strategy, wherein if the interaction event does not meet all the defense distribution strategies, the interaction event is judged to be the effective interaction event.
Optionally, if the interaction event is an effective interaction event, replying to the interaction content in the interaction event through a large language model, and obtaining the replied content includes:
if the interactive event is an effective interactive event, inputting the interactive content in the interactive event into an interactive content input queue with a first preset queue length, wherein the interactive content input queue eliminates the interactive content stored in the interactive content input queue by adopting a least recently used algorithm after the interactive content input queue reaches the first preset queue length;
taking out the interactive content to be processed from the interactive content input queue, and replying the interactive content to be processed through the large language model to obtain replied content;
after the large language model finishes replying to the current interaction content to be processed, taking out the next interaction content to be processed from an interaction content input queue, and inputting the next interaction content to be processed into the large language model until the interaction content does not exist in the interaction content input queue any more.
Optionally, if the interaction event is a valid interaction event, inputting the interaction content in the interaction event into an interaction content input queue having a first preset queue length includes:
if the interaction event is an effective interaction event, inquiring a historical chat record of the audience user in the live broadcast room from a preset historical chat record database according to the user identification information;
splicing the history chat record with the interactive content in the interactive event, and inputting the spliced text data into the interactive content input queue;
the step of taking out the to-be-processed interactive content from the interactive content input queue and replying the to-be-processed interactive content through the large language model, wherein the step of obtaining reply content comprises the following steps:
and taking out the text data to be processed from the interactive content input queue, and replying the text data to be processed through the large language model to obtain replied content.
Optionally, the interactive content input queue includes a plurality of input hierarchical queues with different priorities, and the inputting the interactive content in the interactive event into the interactive content input queue with the preset queue length includes:
Inquiring historical interaction data of the audience user in the live broadcasting room from a preset historical interaction database according to the user identification information;
determining a user representation of the audience user based on the historical interaction data;
determining an input hierarchical queue for storing interactive contents in the interactive event according to the user portrait;
and inputting the interactive contents in the interactive events into the determined input grading queue.
Optionally, if the interaction event is an effective interaction event, replying to the interaction content in the interaction event through a large language model, and after the step of obtaining the replied content, further includes:
judging whether the reply content is effective reply content or not by adopting the defense distribution strategy;
and discarding the reply content if the reply content is invalid.
Optionally, the method further comprises:
and if the reply content is effective reply content, inputting the reply content into a reply content output queue with a second preset queue length, wherein after the reply content output queue reaches the second preset queue length, the reply content stored in the reply content output queue is eliminated by adopting a least recently used algorithm.
Optionally, the reply content output queue includes a plurality of output hierarchical queues with different priorities, and the inputting the reply content into the reply content output queue with the second preset queue length includes:
determining an output staging queue for the reply content input;
and inputting the reply content into the determined output hierarchical queue.
Optionally, the method further comprises:
and acquiring the defense distribution strategy from a preset server.
The application also provides a live broadcast interaction device, live broadcast interaction device includes:
the system comprises an acquisition module, a live broadcast room and a live broadcast room, wherein the acquisition module is used for acquiring an interaction event triggered by a user of a audience in the live broadcast room, wherein the interaction event carries interaction content and user identification information of the user of the audience;
the judging module is used for judging whether the interaction event is an effective interaction event or not by adopting a preset defense distribution strategy;
and the reply module is used for replying the interactive content in the interactive event through a large language model if the interactive event is an effective interactive event, so as to obtain reply content.
The application also provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above method when executing the computer program.
The present application also provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the above method.
In the live broadcasting process by adopting the virtual person, after the interactive event triggered by the audience user in the live broadcasting room is acquired, a preset defense distribution strategy is adopted to judge whether the interactive event is an effective interactive event, so that when the interactive event is judged to be the effective interactive event, the interactive content in the interactive event can be replied through a large language model to obtain replied content, any effective interactive event triggered by the audience user can be replied, interactivity between the virtual person and the audience user is improved, and the user can have better experience. In addition, the interactive events are defended through a defense-setting strategy, so that the number of the interactive events is controlled.
Drawings
Fig. 1 is an application environment schematic diagram of an embodiment of a live interaction method according to an embodiment of the present application;
FIG. 2 is a flow chart of one embodiment of a live interaction method described herein;
FIG. 3 is a detailed schematic diagram of a step of replying to interactive contents in the interactive event through a large language model if the interactive event is an effective interactive event in an embodiment of the present application;
FIG. 4 is a detailed schematic diagram of steps for inputting interactive contents in the interactive event into an interactive content input queue having a first preset queue length if the interactive event is a valid interactive event in an embodiment of the present application;
FIG. 5 is a detailed schematic diagram of a step of inputting interactive contents in the interactive event into an interactive content input queue with a preset queue length according to an embodiment of the application;
FIG. 6 is a flowchart of another embodiment of a live interaction method described herein;
FIG. 7 is a detailed schematic diagram of a step of inputting the reply content into a reply content output queue having a second predetermined queue length according to one embodiment;
FIG. 8 is a block diagram illustrating a live interaction device according to an embodiment of the present disclosure;
fig. 9 is a schematic hardware structure diagram of a computer device for executing a live interaction method according to an embodiment of the present application.
Detailed Description
Advantages of the present application are further described below in conjunction with the drawings and detailed description.
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.
The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure 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 or all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used in this disclosure to describe various information, these 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 disclosure. The word "if" as used herein may be interpreted as "at … …" or "at … …" or "responsive to a determination", depending on the context.
In the description of the present application, it should be understood that the numerical references before the steps do not identify the order of performing the steps, but are only used for convenience in describing the present application and distinguishing each step, and thus should not be construed as limiting the present application.
The following is a term explanation of the present application:
live AI dummy: refers to virtual characters created by artificial intelligence technology for live broadcast and real-time interaction.
Bullet screen: refers to real-time text messages sent by audience users, typically for interaction with the anchor.
Live stream: the transmission of video-on-demand data, which can be transmitted as a steady and continuous stream over a network to a viewer for viewing.
Live broadcast room: in the live broadcast service, each anchor has a unique identifier, and a live broadcast watching page of the anchor can be accessed according to the identifier, namely, a live broadcast room of the anchor;
fig. 1 shows a schematic diagram of an application scenario provided in an embodiment of the present application, where the application scenario includes: server 100, anchor terminal 200, and audience terminal 300. Wherein the server 100, the anchor terminal 200 and the audience terminal 300 are located in a wireless network or a wired network, the server 100 and the anchor terminal 200 and the audience terminal 300 can perform data interaction. In some embodiments, the number of anchor terminals 200 (only one is shown in fig. 1) and the number of audience terminals 300 (only one is shown in fig. 1) may not be limited, and the audience terminals 300 and the anchor terminals 200 may use the server 100 and the CDN (Content Delivery Network ) as transmission intermediaries and implement data interaction with each other through the internet.
In the embodiment of the present application, the server 100 may be a separate server, may be a server cluster, may be a local server, may be a cloud server, and may not be limited in specific server types in the embodiment of the present application. The anchor terminal 200 and the audience terminal 300 may be mobile phones, smart phones, notebook computers, desktop computers, tablet computers, personal digital assistants (Personal Digital Assistant, PDA), media players, smart televisions, wearable electronic devices, etc., and specific types of terminal devices may not be limiting in the embodiments of the present application.
In this embodiment, a live client may be installed in the anchor terminal 200, where the live client is used to implement an anchor to live video. The live client may be a computer Application (APP) installed on the anchor terminal 200, or may be a Web client, which may refer to an Application developed based on a Web architecture.
In this embodiment, a viewer client may be installed in the viewer terminal 300, where the viewer client is configured to receive live content and display the live content. The viewer client may be a computer Application (APP) installed on the viewer terminal 300, or may be a Web client, which may refer to an Application developed based on a Web architecture.
In the following, several embodiments will be provided in the above exemplary application environment to illustrate the live interaction scenario in the present application. The live interaction scheme can be applied to live scenes with live AI virtual persons as the anchor, and can also be applied to live scenes with real users as the anchor. The application is described by taking a live AI virtual person as an example in a live scene of a host.
Fig. 2 is a flow chart of a live interaction method according to an embodiment of the application. The execution main body of the live broadcast interaction method in the application is a live broadcast interaction device. It should be noted that the flowcharts in the method embodiments are not used to limit the order of executing the steps. As can be seen from the figure, the live interaction method provided in this embodiment includes:
step S20, an interaction event triggered by the audience user in the live broadcast room is obtained, wherein the interaction event carries interaction content and user identification information of the audience user.
In particular, the interaction event may include a gift-offering interaction event, a barrage interaction event, a equity song-requesting event, an SC eye-catching message event, a user entering a room event, and the like.
In this embodiment, each interaction event carries the interaction content and user identification information of the audience user. Wherein, the interactive content records the content of the interaction between the audience user and the anchor in the living broadcast room. The user identification information is information for distinguishing different audience users, and the user identification information can be user ID.
As an example, when the interaction event is a gift-offering interaction event, the interaction content may be "XX gives a rose"; when the interaction event is a barrage interaction event, the interaction content may be "XXX"; when the interactive event is a rights song-requesting event, the interactive content can be "XX song is requested by XX"; when the interaction event is an SC striking message event, the interaction content can leave a XX message for XX; when the interactive event is a room entrance event, the interactive content may be "XX enters a living broadcast room".
And S21, judging whether the interaction event is an effective interaction event by adopting a preset defense strategy.
Specifically, the arming policy is a policy for determining whether an interaction event is a valid interaction event. The arming policy may include one or more.
In an embodiment, the defense strategy may be a semantic defense strategy, an interactive quota defense strategy, a blacklist defense strategy, a hit wind control defense strategy, and the like.
The semantic defense distribution strategy is a strategy for conducting defense distribution according to semantics. When the semantic defense strategy judges that the semantics of the currently acquired interaction event are the same as those of the interaction event acquired before the current moment, the semantic defense strategy judges that the currently acquired interaction event needs to be defended, namely the currently acquired interaction event needs to be discarded. When the semantic defense strategy judges that the semantics of the currently acquired interaction event are different from those of the interaction event acquired before the current moment, the semantic defense strategy does not defend the currently acquired interaction event, namely the currently acquired interaction event is considered to be an effective interaction event.
The interactive quota distribution strategy is a strategy for distributing according to the quota amount of unit time of an interactive event. When the interactive quota distribution strategy determines that the number of the interactive events acquired in the current unit time exceeds the unit time quota, the interactive events acquired in the unit time are distributed, namely the interactive events acquired in the unit time are required to be discarded. When the interactive quota distribution strategy judges that the number of the interactive events acquired in the current unit time is smaller than the number of the unit time quota, the current acquired interactive events are not distributed, namely the current acquired interactive events are considered to be effective interactive events. The limit amount of unit time can be set and adjusted according to practical situations, for example, the limit amount of unit time is 3 interaction events per minute.
The blacklist defense arrangement strategy is a strategy for conducting defense arrangement according to a blacklist database. When the blacklist defense arrangement strategy judges that the currently acquired interaction event belongs to the interaction event triggered by the blacklist user, the interaction event is defended, namely the currently acquired interaction event is discarded. And when the blacklist defense arrangement strategy judges that the currently acquired interaction event does not belong to the interaction event triggered by the blacklist user, the currently acquired interaction event is determined to belong to the effective interaction event. And the blacklist database records the user identification information of all blacklist users.
The hit wind control defense arrangement strategy is a strategy for conducting defense arrangement according to a wind control database. When the hit wind control defense arrangement strategy judges that the currently acquired interaction event contains wind control words, the interaction event is defended, namely the currently acquired interaction event is discarded. When the hit wind control defense strategy judges that the currently acquired interaction event does not contain wind control words, the currently acquired interaction event is determined to belong to an effective interaction event. Wherein, wind control words which need wind control are recorded in the wind control database.
In an exemplary embodiment, when the defense strategy is adopted to determine that the interaction event is an invalid interaction event, the interaction event is discarded.
In this embodiment, after an interaction event is acquired, when a defense deployment policy is adopted to determine whether the interaction event is an invalid interaction event, and when the interaction event is determined to be an invalid interaction event, the interaction event is discarded, so that the number of subsequent interaction content to be replied through a large language model is reduced.
In an exemplary embodiment, when the defense strategy includes at least one defense strategy, the determining whether the interaction event is a valid interaction event using a preset defense strategy includes: and judging whether the interaction event is an effective interaction event by adopting the at least one defense distribution strategy, wherein if the interaction event does not meet all the defense distribution strategies, the interaction event is judged to be the effective interaction event.
Specifically, when the defense strategy is only one, the defense strategy is adopted to judge whether the interaction event is a valid interaction event. When the interactivity event does not satisfy such a defense strategy, then the interactivity event may be determined to be a valid interactivity event.
When the distribution strategies are multiple, multiple distribution strategies are adopted to judge whether the interaction event is a valid interaction event. And when the interaction event does not meet all the defense distribution strategies, judging that the interaction event is a valid interaction event.
In this embodiment, by adopting various defense strategies to determine whether the interaction event is a valid interaction event, the number of subsequent replies to the interaction content through the large language model can be further reduced.
In an exemplary embodiment, in order to facilitate the flexible configuration of the defense deployment policy, the defense deployment policy may be obtained from a preset server.
Specifically, the user may configure a defense distribution policy for determining whether the interaction event is a valid interaction event in the server in advance, and then, when the defense distribution policy needs to be adopted to determine whether the interaction event is a valid interaction event, the defense distribution policy may be obtained from the server first, and then, the obtained defense distribution policy is adopted to determine whether the interaction event is a valid interaction event.
When the user wants to change the distribution strategy, the user can change the distribution strategy in the server, and after the distribution strategy is changed, the changed distribution strategy is pushed to the live interaction device.
And S22, if the interaction event is an effective interaction event, replying the interaction content in the interaction event through a large language model to obtain replied content.
Specifically, when the interactive event is determined to be a valid interactive event, the interactive content in the interactive event is input into the large language model, so that the large language model can answer to the interactive content.
The large language model (Large Language Model, LLM) is a natural language processing model constructed by using a deep learning technology, can process a large amount of text data, and has strong language understanding and generating capability. It uses billions of parameters to train, understand semantics, context, and generate natural language responses. Large language models achieve significant achievements in a number of fields, such as natural language understanding, dialog systems, text summaries, etc.
In this embodiment, after obtaining the reply content, the reply content may be converted into a voice broadcast in the live broadcast room.
In the live broadcast process of the anchor (live anchor or virtual anchor), after the interactive event triggered by the audience user in the live broadcast room is acquired, a preset defense arrangement strategy is adopted to judge whether the interactive event is an effective interactive event, so that when the interactive event is judged to be the effective interactive event, the interactive content in the interactive event can be replied through a large language model to obtain replied content, any effective interactive event triggered by the audience user can be replied, interactivity between the virtual character and the audience user is improved, and the user can have better experience. In addition, the interactive events are defended through a defense-setting strategy, so that the number of the interactive events is controlled.
In an exemplary embodiment, referring to fig. 3, if the interaction event is a valid interaction event, replying to the interaction content in the interaction event through a large language model, and obtaining reply content includes:
step S30, if the interactive event is an effective interactive event, inputting the interactive content in the interactive event into an interactive content input queue with a first preset queue length, wherein the interactive content input queue eliminates the interactive content stored in the interactive content input queue by adopting a least recently used algorithm after the interactive content input queue reaches the first preset queue length.
In particular, during a live anchor, there may be multiple audience users triggering interaction events at the same time. However, the large language model requires a certain time for processing each interactive content, and thus, in order to allow the interactive content to be replied to by the large language model. In this embodiment, when it is determined that the current interaction event is a valid interaction event, the interaction content of the interaction event is input into the interaction content input queue.
The interactive content input queue is a queue for storing interactive content. The interactive content input queue has a first preset queue length. The first preset queue length may be set according to an actual situation, for example, if the first preset queue length is 6, it indicates that the interactive content input queue may store at most 6 interactive contents.
In this embodiment, in order to facilitate flexible setting of the first preset queue length, the first preset queue length may be set and modified in the server. And when the first preset queue length needs to be acquired, acquiring the first preset queue length from a server.
In this embodiment, since the large language model processing is limited by available resources, interactive contents that can be processed in a unit time are limited, and interactive contents input into the interactive contents input queue in a unit time may exceed the large language model processing capability. Therefore, in this embodiment, after the length of the interactive content input queue reaches the first preset queue length, if there is interactive content to be input into the interactive content input queue, the interactive content that is stored in the interactive content input queue and is not used most recently is eliminated by using a least recently used (Least Recently Used, LRU) algorithm, that is, the interactive content that is stored in the interactive content input queue and is not used most recently is discarded first, and then the interactive content that is required to be input into the interactive content input queue is input into the interactive content input queue.
In this embodiment, the interactive content stored in the interactive content input queue is eliminated by using the LRU algorithm, so that the interactive content stored in the interactive content input queue is dynamically updated in real time, and each piece of interactive content is ensured to be the interactive content which meets the expected latest need to be processed.
And S31, taking out the to-be-processed interactive content from the interactive content input queue, and replying the to-be-processed interactive content through the large language model to obtain replied content.
Specifically, when the large language model is required to answer the interactive content, the interactive content to be processed is firstly taken out from the interactive content input queue, and then the taken interactive content is input into the large language model. After receiving the interaction content to be processed, the large language model replies the interaction content to be processed, so that reply content is obtained.
It will be appreciated that if the large language model can process N pieces of interactive content simultaneously at a time, N pieces of interactive content are taken out of the interactive content input queue, and then input into the large language model, so as to reply to the N pieces of interactive content in parallel through the large language model.
Step S32, after the large language model finishes replying to the current interaction content to be processed, taking out the next interaction content to be processed from the interaction content input queue, and inputting the next interaction content to be processed into the large language model until the interaction content does not exist in the interaction content input queue any more.
Specifically, in order to make the large language model answer to the interactive content in time, in this embodiment, when the large language model answers to the current interactive content to be processed, the answer state of the large language model to the interactive content is obtained in real time, if the answer state of the large language model to the current interactive content is that the answer is completed, it indicates that the large language model has an idle resource to process the new interactive content to be processed, at this time, the next piece of interactive content to be processed is immediately fetched from the interactive content input queue, and is input into the large language model until the interactive content no longer exists in the interactive content input queue, so that the large language model can immediately answer to the next piece of interactive content to be processed.
According to the embodiment, after the interactive event is judged to be effective, the interactive event is stored in the interactive content input queue with the LRU elimination strategy, so that each piece of interactive content processed by the large language model is the interactive content which is expected to be processed up to date, and the latest interactive content can be replied in time, and better interactive experience is provided for a user.
In an exemplary embodiment, referring to fig. 4, if the interaction event is a valid interaction event, inputting the interaction content in the interaction event into the interaction content input queue having the first preset queue length includes:
step S40, if the interaction event is an effective interaction event, inquiring a historical chat record of the audience user in the live broadcast room from a preset historical chat record database according to the user identification information.
Specifically, the historical chat record database is used for recording chat records of each audience user with each anchor in the live broadcast room in a historical time period. The history period may be set according to actual conditions, for example, the history period is a past week, a past month, or the like. The chat record refers to a record that a spectator interacts with a host in a live broadcasting room by sending a barrage, namely, a barrage sent by a spectator user and a reply message of the host to the spectator user serve as a chat record.
In this embodiment, after the interaction event is determined to be a valid interaction event, all the historical chat records matching the user identification information are queried from the historical chat record database.
It will be appreciated that all historical chat records queried include chat records generated by users interacting with the anchor in various live rooms. Thus, after all matching historical chat records are queried, the historical chat records of the audience user and the anchor in the live broadcast room are screened out from the historical chat records.
It will be appreciated that in order to facilitate querying of the historical chat log of the viewer user in the live broadcast room, the chat log may be stored with user identification information and live broadcast room identification information as key values when the chat log is stored in the historical chat log database.
And step S41, splicing the history chat record with the interactive content in the interactive event, and inputting the spliced text data into the interactive content input queue.
Specifically, after the history chat record is obtained, the history chat record is spliced with the interactive content in the interactive event, so that the text data obtained by splicing can contain global context of the chat of the audience user, and more accurate reply content can be obtained when the interactive content is replied through a large language model later.
In this embodiment, when the splicing is performed, the interactive content may be spliced at the rearmost of the content chat record.
Correspondingly, the step of taking out the to-be-processed interactive content from the interactive content input queue and replying the to-be-processed interactive content through the large language model, and the step of obtaining reply content comprises the following steps: and taking out the text data to be processed from the interactive content input queue, and replying the text data to be processed through the large language model to obtain replied content.
Specifically, since the text data input into the large language model contains the global context of chat with the anchor by the audience, a more accurate reply content can be generated when the text data to be processed is replied by the large language model.
In this embodiment, the large language model is used to answer the interactive content including the chat record of the audience user and the anchor, so that the accuracy of the generated answer content can be improved.
In an exemplary embodiment, to allow for preferential reply to important interactive content, in this embodiment, the interactive content input queue includes a plurality of input staging queues having different priorities. Different input staging queues have different processing priorities of the interactive contents stored therein, and the higher the priority is, the higher the processing priority of the interactive contents stored therein is.
As an example, the interactive contents input queue includes a priority queue having a 1 st priority level, a gift queue having a 2 nd priority level, a barrage queue having a 3 rd priority level, a song requesting queue having a 4 th priority level, a clear message queue having a 5 th priority level, a house entering queue having a 6 th priority level, and the like. The priority of the 1 st priority level is higher than the 2 nd priority level, so that the 5 th priority level is higher than the 6 th priority level.
Referring to fig. 5, the inputting the interactive contents in the interactive event into the interactive contents input queue with the preset queue length includes:
and step S50, inquiring historical interaction data of the audience user in the live broadcasting room from a preset historical interaction database according to the user identification information.
Specifically, the historical interaction database is used for recording interaction data of each audience user and each anchor in a live broadcast room in a historical time period. The history period may be set according to actual conditions, for example, the history period is a past week, a past month, or the like. The interactive data refers to data for recording an interactive event triggered by a viewer user in a live broadcast room, namely, the interactive event is recorded every time the viewer user triggers an interactive event in the live broadcast room. The interactive data may record only the type of the interactive event, and does not need to record the interactive content in the interactive event. For example, if the current interaction event is a barrage interaction event, the interaction data may be a barrage interaction event.
In this embodiment, after the interaction event is determined to be an effective interaction event, all the historical interaction data matched with the user identification information is queried from the historical interaction database.
It will be appreciated that all of the historical interaction data queried includes the historical interaction data of the user in each live room. Therefore, after all the matched historical interaction data are queried, the historical interaction data of the audience user in the live broadcast room are screened out from the historical interaction data.
It will be appreciated that in order to facilitate querying of the viewer user's historical interaction data in the live broadcast room, the interaction data may be stored with the user identification information and live broadcast room identification information as key values when storing the interaction data in the historical interaction database.
And S51, determining the user portrait of the audience user according to the historical interaction data.
Specifically, the user portrayal is a label describing the user characteristics of the viewer user.
In this embodiment, the historical interaction data may be analyzed using an existing user portrayal algorithm to obtain a user portrayal of the audience user.
In a particular scenario, the tag may be active, inactive, a paid user, a non-paid user, etc.
Step S52, determining an input grading queue for storing the interactive contents in the interactive event according to the user portrait.
Specifically, a mapping relation table of different user portraits and different input grading queues can be pre-established, so that after the user portraits are obtained, the input grading queues of the interactive contents in the interactive events can be determined by querying the mapping relation table.
In another embodiment, when the interactive contents input queue includes a priority queue having a 1 st priority level, a gift queue having a 2 nd priority level, a barrage queue having a 3 rd priority level, a song requesting queue having a 4 th priority level, a eye-catching message queue having a 5 th priority level, a house entering queue having a 6 th priority level, etc., the user portrait may also be used only to determine whether the interactive contents of the interactive event should be stored to the priority queue having the 1 st priority level. If it is determined from the user profile that the interactive content of the interactive event should be stored in the priority queue having the 1 st priority, then the interactive content of the interactive event is subsequently stored in the priority queue having the 1 st priority. If it is determined from the user profile that the interactive content of the interactive event should not be stored in the priority queue having the priority 1, the interactive event is continuously stored in the corresponding input ranking queue according to the type of the interactive event. For example, if the interactive event is a song requesting event, the interactive content in the interactive event is stored in a song requesting queue with the 4 th priority level.
And step S53, inputting the interactive contents in the interactive event into the determined input grading queue.
Specifically, after the input rank queue in which the interactive content should be stored is determined, the interactive content is stored in the determined input rank queue.
It can be understood that each input queue stage also has the first preset queue length, and after the length reaches the first preset queue length, the interactive content stored in the input queue stage is eliminated by using the least recently used algorithm.
In this embodiment, the interactive content input queue is divided into a plurality of input hierarchical queues with different priorities, so that when the large language model processes the interactive content, the processing sequence of each interactive content can be adjusted according to the user image, and better interactive experience can be provided for high-value users.
In an exemplary embodiment, referring to fig. 6, if the interaction event is a valid interaction event, the step of replying to the interaction content in the interaction event through the large language model to obtain the replied content further includes:
and step S60, judging whether the reply content is valid reply content or not by adopting the defense strategy.
Specifically, since the interaction amount between the anchor and the audience is limited in unit time, in this embodiment, after the reply content is obtained, the reply content may be screened by using the defense strategy.
The defense strategy is described in detail in the above embodiments, and will not be described in detail in this embodiment.
Step S61, if the reply content is invalid, discarding the reply content.
Specifically, when the current reply content is judged to be invalid, the reply content is directly discarded, so that the reply content is not required to be subjected to voice broadcasting in the follow-up process.
In this embodiment, by discarding the invalid reply content, the reply content of the subsequent voice broadcast can be reduced.
In an exemplary embodiment, the method further comprises:
and if the reply content is effective reply content, inputting the reply content into a reply content output queue with a second preset queue length, wherein after the reply content output queue reaches the second preset queue length, the reply content stored in the reply content output queue is eliminated by adopting a least recently used algorithm.
Specifically, the reply content output queue is a queue for storing reply content.
In this embodiment, since a certain time is required when the host (the live host or the virtual host) plays each piece of reply content, in order to enable the effective reply content to be voice-played, after the effective reply content is obtained, the reply content is input into a reply content output queue with a second preset queue length for caching. Thus, when the reply content needs to be broadcast later, the reply content can be taken out from the reply content output queue for voice broadcast.
In order to enable the latest reply content to be subjected to voice broadcasting, in this embodiment, the length of a queue in a reply content output queue may be limited, and after the length reaches the second preset queue length, the reply content stored in the reply content output queue is eliminated by adopting a least recently used algorithm, so that the reply content stored in the reply content output queue is dynamically updated in real time, and each piece of reply content can be ensured to be the reply content meeting the expected latest requirement for voice broadcasting.
The second preset queue length may be set according to an actual situation, for example, if the second preset queue length is 6, it indicates that the reply content output queue may store at most 6 pieces of reply content.
In this embodiment, after the reply content is determined to be valid reply content, the reply content is stored in the reply content output queue with the LRU elimination policy, so that each piece of broadcasted reply content is the reply content meeting the expected latest requirement for broadcasting when the reply content is broadcast in the follow-up process, and the interactive experience of the user is improved.
In an exemplary embodiment, the reply content output queue includes a plurality of output staging queues having different priorities. Wherein the processing priorities of the reply contents stored in the different output hierarchical queues are different, and the higher the priority is, the higher the processing priority of the reply contents stored in the different output hierarchical queues is.
In one embodiment, the reply content output queue includes the same number of output rank queues as the number of input rank queues included in the interactive content input queue, and each output rank queue corresponds to one input rank queue.
As an example, the reply content output queue also includes a priority queue with a 1 st priority level, a gift queue with a 2 nd priority level, a barrage queue with a 3 rd priority level, a song requesting queue with a 4 th priority level, a clear message queue with a 5 th priority level, a room entering queue with a 6 th priority level, and so on. The priority of the 1 st priority level is higher than the 2 nd priority level, so that the 5 th priority level is higher than the 6 th priority level.
Referring to fig. 7, the inputting the reply content into the reply content output queue having the second preset queue length includes:
step S70, determining an output rank queue of the reply content input.
Specifically, since the reply content corresponds to the interactive content, in this embodiment, when determining the output rank queue of the reply content, the priority of the interactive content corresponding to the reply content may be used as the priority of the reply content, and then the output rank queue may be determined according to the determined priority.
As an example, the interactive content corresponding to the reply content is stored in the song requesting queue with the 4 th priority level, and the output hierarchical queue corresponding to the reply content is also the song requesting queue with the 4 th priority level.
Step S71, inputting the reply content into the determined output hierarchical queue.
Specifically, after the output rank queue is determined, the reply content is input to the determined output rank queue.
It can be understood that each output queue stage described above also has the second preset queue length, and after the length reaches the second preset queue length, the reply content stored in the output queue stage is eliminated by using the least recently used algorithm.
In this embodiment, the reply content output queue is divided into a plurality of output hierarchical queues with different priorities, so that when the reply content is subsequently broadcast in a voice manner, the reply content with a high priority can be broadcast in a priority manner. In addition, by setting up a plurality of output rank queues, it is possible to ensure that interactive reply content output can be always provided.
Referring to fig. 8, a block diagram of a live interaction device 80 according to an embodiment of the present application is shown.
In this embodiment, the live interaction device 80 includes a series of computer program instructions stored on a memory, which when executed by a processor, implement the live interaction functions of the embodiments of the present application. In some embodiments, based on the specific operations implemented by the portions of the computer program instructions, the live interaction device 80 may be divided into one or more modules, which may be specifically divided as follows:
The acquiring module 81 is configured to acquire an interaction event triggered by a viewer user in a live broadcast room, where the interaction event carries interaction content and user identification information of the viewer user;
a judging module 82, configured to judge whether the interaction event is a valid interaction event by adopting a preset defense deployment policy;
and a reply module 83, configured to reply to the interactive content in the interactive event through a large language model if the interactive event is a valid interactive event, so as to obtain reply content.
In an exemplary embodiment, the live interaction device 80 further includes a discard module.
The discarding module is configured to discard the interaction event if the interaction event is an invalid interaction event.
In an exemplary embodiment, the defense distribution policy includes at least one defense distribution policy, and the determining module 82 is further configured to determine whether the interaction event is a valid interaction event by using the at least one defense distribution policy, where if the interaction event does not satisfy all the defense distribution policies, determine that the interaction event is a valid interaction event.
In an exemplary embodiment, the reply module 83 is further configured to, if the interaction event is a valid interaction event, input the interaction content in the interaction event into an interaction content input queue having a first preset queue length, where after the length of the interaction content input queue reaches the first preset queue length, reject the interaction content stored in the interaction content input queue by using a least recently used algorithm; taking out the interactive content to be processed from the interactive content input queue, and replying the interactive content to be processed through the large language model to obtain replied content; after the large language model finishes replying to the current interaction content to be processed, taking out the next interaction content to be processed from an interaction content input queue, and inputting the next interaction content to be processed into the large language model until the interaction content does not exist in the interaction content input queue any more.
In an exemplary embodiment, the reply module 83 is further configured to query a preset historical chat record database for the audience user in the live broadcast room according to the user identification information if the interaction event is a valid interaction event; splicing the history chat record with the interactive content in the interactive event, and inputting the spliced text data into the interactive content input queue;
the reply module 83 is further configured to take out text data to be processed from the interactive content input queue, and reply the text data to be processed through the large language model, so as to obtain reply content.
In an exemplary embodiment, the interactive content input queue includes a plurality of input hierarchical queues with different priorities, and the reply module 83 is further configured to query a preset historical interaction database for historical interaction data of the audience user in the live broadcast room according to the user identification information; determining a user representation of the audience user based on the historical interaction data; determining an input hierarchical queue for storing interactive contents in the interactive event according to the user portrait; and inputting the interactive contents in the interactive events into the determined input grading queue.
In an exemplary embodiment, the determining module 82 is further configured to determine whether the reply content is valid reply content using the defense deployment policy;
the discarding module is further configured to discard the reply content if the reply content is invalid.
In an exemplary embodiment, the live interaction device 80 further includes an input module.
The input module is configured to input the reply content into a reply content output queue having a second preset queue length if the reply content is valid, where the reply content output queue eliminates the reply content stored in the reply content output queue by using a least recently used algorithm after the reply content output queue has a length up to the second preset queue length.
In an exemplary embodiment, the reply content output queue includes a plurality of output rank queues having different priorities, and the input module is further configured to determine an output rank queue for the reply content input; and inputting the reply content into the determined output hierarchical queue.
In an exemplary embodiment, the obtaining module 81 is further configured to obtain the defense deployment policy from a preset server.
Fig. 9 schematically shows a hardware architecture diagram of a computer device 9 adapted to implement a live interaction method according to an embodiment of the present application. In the present embodiment, the computer device 9 is a device capable of automatically performing numerical calculation and/or information processing in accordance with instructions set or stored in advance. As shown in fig. 9, the computer device 9 includes at least, but is not limited to: memory 120, processor 121, and network interface 122 may be communicatively linked to each other by a system bus. Wherein:
memory 120 includes at least one type of computer-readable storage medium that may be volatile or nonvolatile, and specifically, readable storage media include flash memory, hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), random Access Memory (RAM), static Random Access Memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 120 may be an internal storage module of the computer device 9, such as a hard disk or memory of the computer device 9. In other embodiments, the memory 120 may also be an external storage device of the computer device 9, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 9. Of course, the memory 120 may also include both internal memory modules of the computer device 9 and external memory devices. In this embodiment, the memory 120 is generally used to store an operating system and various application software installed on the computer device 9, such as program codes of a live interaction method. In addition, the memory 120 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 121 may be a central processing unit (Central Processing Unit, simply CPU), controller, microcontroller, microprocessor, or other live interaction chip in some embodiments. The processor 121 is typically used to control the overall operation of the computer device 9, such as performing control and processing related to data interaction or communication with the computer device 9, and the like. In this embodiment, the processor 121 is configured to execute program codes or process data stored in the memory 120.
The network interface 122 may comprise a wireless network interface or a wired network interface, which network interface 122 is typically used to establish a communication link between the computer device 9 and other computer devices. For example, the network interface 122 is used to connect the computer device 9 to an external terminal through a network, establish a data transmission channel and a communication link between the computer device 9 and the external terminal, and the like. The network may be a wireless or wired network such as an Intranet (Intranet), the Internet (Internet), a global system for mobile communications (Global System of Mobile communication, abbreviated as GSM), wideband code division multiple access (Wideband Code Division Multiple Access, abbreviated as WCDMA), a 4G network, a 5G network, bluetooth (Bluetooth), wi-Fi, etc.
It should be noted that fig. 9 only shows a computer device having components 120-122, but it should be understood that not all of the illustrated components are required to be implemented, and that more or fewer components may be implemented instead.
In this embodiment, the live interaction method stored in the memory 120 may be divided into one or more program modules and executed by one or more processors (the processor 121 in this embodiment) to complete the present application.
The present application provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor, implements the steps of the live interaction method in the embodiments.
In this embodiment, the computer-readable storage medium includes a flash memory, a hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the computer readable storage medium may be an internal storage unit of a computer device, such as a hard disk or a memory of the computer device. In other embodiments, the computer readable storage medium may also be an external storage device of a computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), etc. that are provided on the computer device. Of course, the computer-readable storage medium may also include both internal storage units of a computer device and external storage devices. In this embodiment, the computer readable storage medium is typically used to store an operating system and various application software installed on the computer device, such as program codes of the live interaction method in the embodiment. Furthermore, the computer-readable storage medium may also be used to temporarily store various types of data that have been output or are to be output.
The apparatus embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over at least two network elements. Some or all modules in the system can be screened out according to actual needs to achieve the purpose of the embodiment of the application. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus a general purpose hardware platform, or may be implemented by hardware. Those skilled in the art will appreciate that all or part of the processes implementing the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, where the program may be stored in a computer readable storage medium, and where the program may include processes implementing the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), or the like.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application.
Claims (13)
1. A live interaction method, the method comprising:
acquiring an interaction event triggered by a viewer user in a live broadcast room, wherein the interaction event carries interaction content and user identification information of the viewer user;
judging whether the interaction event is an effective interaction event by adopting a preset defense distribution strategy;
and if the interaction event is an effective interaction event, replying the interaction content in the interaction event through a large language model to obtain replied content.
2. The live interaction method of claim 1, further comprising:
and discarding the interaction event if the interaction event is an invalid interaction event.
3. The live interaction method according to claim 1, wherein the defense deployment strategy includes at least one defense deployment strategy, and the determining whether the interaction event is a valid interaction event using a preset defense deployment strategy includes:
and judging whether the interaction event is an effective interaction event by adopting the at least one defense distribution strategy, wherein if the interaction event does not meet all the defense distribution strategies, the interaction event is judged to be the effective interaction event.
4. The live interaction method according to claim 1, wherein if the interaction event is a valid interaction event, replying to the interaction content in the interaction event through a large language model, and obtaining the reply content includes:
if the interactive event is an effective interactive event, inputting the interactive content in the interactive event into an interactive content input queue with a first preset queue length, wherein the interactive content input queue eliminates the interactive content stored in the interactive content input queue by adopting a least recently used algorithm after the interactive content input queue reaches the first preset queue length;
taking out the interactive content to be processed from the interactive content input queue, and replying the interactive content to be processed through the large language model to obtain replied content;
After the large language model finishes replying to the current interaction content to be processed, taking out the next interaction content to be processed from an interaction content input queue, and inputting the next interaction content to be processed into the large language model until the interaction content does not exist in the interaction content input queue any more.
5. The live interaction method according to claim 4, wherein if the interaction event is a valid interaction event, the inputting the interaction content in the interaction event into the interaction content input queue having the first preset queue length includes:
if the interaction event is an effective interaction event, inquiring a historical chat record of the audience user in the live broadcast room from a preset historical chat record database according to the user identification information;
splicing the history chat record with the interactive content in the interactive event, and inputting the spliced text data into the interactive content input queue;
the step of taking out the to-be-processed interactive content from the interactive content input queue and replying the to-be-processed interactive content through the large language model, wherein the step of obtaining reply content comprises the following steps:
And taking out the text data to be processed from the interactive content input queue, and replying the text data to be processed through the large language model to obtain replied content.
6. The live interaction method of claim 4, wherein the interactive content input queue comprises a plurality of input hierarchical queues having different priorities, and wherein inputting the interactive content in the interactive event into the interactive content input queue having a preset queue length comprises:
inquiring historical interaction data of the audience user in the live broadcasting room from a preset historical interaction database according to the user identification information;
determining a user representation of the audience user based on the historical interaction data;
determining an input hierarchical queue for storing interactive contents in the interactive event according to the user portrait;
and inputting the interactive contents in the interactive events into the determined input grading queue.
7. The live interaction method according to any one of claims 1 to 6, wherein if the interaction event is a valid interaction event, replying to the interaction content in the interaction event through a large language model, and after the step of obtaining the replied content, further comprising:
Judging whether the reply content is effective reply content or not by adopting the defense distribution strategy;
and discarding the reply content if the reply content is invalid.
8. The live interaction method of claim 7, wherein the method further comprises:
and if the reply content is effective reply content, inputting the reply content into a reply content output queue with a second preset queue length, wherein after the reply content output queue reaches the second preset queue length, the reply content stored in the reply content output queue is eliminated by adopting a least recently used algorithm.
9. The live interaction method of claim 8, wherein the reply content output queue comprises a plurality of output staging queues having different priorities, and wherein inputting the reply content into the reply content output queue having a second preset queue length comprises:
determining an output staging queue for the reply content input;
and inputting the reply content into the determined output hierarchical queue.
10. The live interaction method of claim 1, further comprising:
And acquiring the defense distribution strategy from a preset server.
11. A live interaction device, characterized in that the live interaction device comprises:
the system comprises an acquisition module, a live broadcast room and a live broadcast room, wherein the acquisition module is used for acquiring an interaction event triggered by a user of a audience in the live broadcast room, wherein the interaction event carries interaction content and user identification information of the user of the audience;
the judging module is used for judging whether the interaction event is an effective interaction event or not by adopting a preset defense distribution strategy;
and the reply module is used for replying the interactive content in the interactive event through a large language model if the interactive event is an effective interactive event, so as to obtain reply content.
12. A computer device, characterized in that it comprises a memory, a processor and a computer program stored on the memory and executable on the processor, which processor implements the steps of the method according to any one of claims 1 to 10 when the computer program is executed.
13. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311750380.6A CN117714804A (en) | 2023-12-18 | 2023-12-18 | Live interaction method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311750380.6A CN117714804A (en) | 2023-12-18 | 2023-12-18 | Live interaction method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117714804A true CN117714804A (en) | 2024-03-15 |
Family
ID=90156653
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311750380.6A Pending CN117714804A (en) | 2023-12-18 | 2023-12-18 | Live interaction method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117714804A (en) |
-
2023
- 2023-12-18 CN CN202311750380.6A patent/CN117714804A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10210342B2 (en) | Centralized throttling service | |
| US9686329B2 (en) | Method and apparatus for displaying webcast rooms | |
| WO2014183427A1 (en) | Method and apparatus for displaying webcast rooms | |
| CN107528818B (en) | Data processing method and device for media file | |
| CN110659121A (en) | Task data acquisition method and device, task configuration method and device and server | |
| US11388561B2 (en) | Providing a summary of media content to a communication device | |
| CN106487655B (en) | Message interaction method and device and processing server | |
| WO2015027907A1 (en) | Methods and systems for visiting user groups | |
| US20160203144A1 (en) | Method and System for Processing Associated Content | |
| CN113626624B (en) | Resource identification method and related device | |
| CN109657164B (en) | Method, device and storage medium for publishing message | |
| CN113315981B (en) | Task data updating method, device and system, electronic equipment and storage medium | |
| CN112312145A (en) | Access server, burst traffic caching method, system, computer device and readable storage medium | |
| CN106487653B (en) | Message processing method and server | |
| US20170171333A1 (en) | Method and electronic device for information pushing | |
| CN111782919A (en) | Online document processing method and device, computer equipment and storage medium | |
| CN117714804A (en) | Live interaction method and device, computer equipment and storage medium | |
| CN113515336B (en) | Live room joining method, creation method, device, equipment and storage medium | |
| CN111405348B (en) | Method, apparatus and computer storage medium for media playing | |
| CN115017538A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN112491939A (en) | Multimedia resource scheduling method and system | |
| CN107222559B (en) | Information calling method | |
| CN115119020B (en) | Live broadcasting room interaction control method, device, electronic equipment, medium and program product | |
| CN115334154B (en) | Page access method, page access device, computer equipment and storage medium | |
| CN108810043B (en) | Method and device for storing comment information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |