Information processing method and device, equipment and medium

Abstract

The invention provides an information processing method and device, equipment and a medium, and the method is applied to a first server, and comprises the steps: pulling a target AI virtual terminal into a video conference; sending a video stream of a participant to the target AI virtual terminal through an AI server, so that the AI virtual terminal identifies the video stream to obtain a picture of the participant; an operation instruction returned by the AI server is received, and the operation instruction is generated when the AI server detects the picture of the conference participant and detects that a specified action exists in the picture of the conference participant; and controlling the speaking of the participant according to the operation instruction so as to solve the problem that the speaker cannot be switched when the terminal to which the chairman belongs fails at present.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to an information processing method, device, equipment and medium. Background Art

[0002] With the rapid development of network technology, two-way communications such as video conferencing and video teaching have become widely popular in users' lives, work, and study.

[0003] Existing video conferences usually include multiple participants and one chairperson. The chairperson performs operations such as switching speakers on the video conference. If the chairman's terminal fails, the video conference will be out of control and operations such as switching speakers cannot be performed. Summary of the invention

[0004] The present invention provides an information processing method, device, equipment and medium to solve the problem that the speaker cannot be switched when a terminal belonging to the chairman fails.

[0005] In order to solve the above problem, the present invention discloses an information processing method, which is applied to a first server and includes:

[0006] Pull the target AI virtual terminal into the video conference;

[0007] Sending the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant;

[0008] receiving an operation instruction returned by the AI ​​server, where the operation instruction is generated when the AI ​​server detects the participant's picture and detects that a specified action exists in the participant's picture;

[0009] The speech of the conference participant is controlled according to the operation instruction.

[0010] Optionally, before the target AI virtual terminal is pulled into the microphone grabbing conference, the method further includes:

[0011] Receive AI microphone grabbing conference request triggered by client;

[0012] The first server selects a target AI virtual terminal from the AI ​​virtual terminals of the connected AI servers in response to the AI ​​microphone grabbing conference request;

[0013] A response is sent to the client indicating that the conference is successfully started.

[0014] Optionally, the AI ​​microphone grabbing conference request carries a conference identifier and a main message number of the AI ​​server;

[0015] The first server selects a target AI virtual terminal from the AI ​​virtual terminals of the connected AI server in response to the AI ​​microphone grabbing conference request, including:

[0016] The first server responds to the AI ​​microphone grabbing conference request, selects an AI virtual terminal with idle status information from the AI ​​virtual terminals of the connected AI server according to the conference identifier and the main message number, and uses the idle AI virtual terminal as the target AI virtual terminal.

[0017] Optionally, controlling the speech of the participant according to the operation instruction includes:

[0018] Parsing the operation instruction to obtain a specified action identifier and a channel number;

[0019] Determine the participant from the corresponding relationship between the participant and the AI ​​switching channel number according to the channel number;

[0020] The speech of the conference participant is controlled according to the designated action identifier.

[0021] Optionally, the method further comprises:

[0022] When receiving the AI ​​microphone grabbing conference request to close the video conference, the first server removes the target AI virtual terminal from the video conference and notifies the AI ​​server to unbind the target AI virtual terminal.

[0023] In order to solve the above problems, the present invention further discloses an information processing method, which is applied to an AI server and includes:

[0024] Receiving a video stream of a conference participant sent by the first server;

[0025] Sending the video stream to the AI ​​virtual terminal so that the AI ​​virtual terminal recognizes the video stream and obtains the picture of the participant;

[0026] Detecting the participant's picture, and generating an operation instruction when a specified action is detected in the participant's picture;

[0027] The operation instruction is sent to the first server, so that the first server controls the speech of the participant according to the operation instruction.

[0028] Optionally, before receiving the video stream of the conference participant sent by the first server, the method further includes:

[0029] Configuring an AI virtual terminal on the AI ​​server;

[0030] When receiving an AI microphone grabbing conference request triggered by a client, a target AI virtual terminal is selected from the configured AI virtual terminals.

[0031] In order to solve the above problem, the present invention further discloses an information processing device, which is applied to a first server and includes:

[0032] The conference module is used to connect the target AI virtual terminal to the video conference;

[0033] A sending module, used to send the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant;

[0034] A return module, configured to receive an operation instruction returned by the AI ​​server, wherein the operation instruction is generated when the AI ​​server detects the participant's picture and detects that a specified action exists in the participant's picture;

[0035] The first control module is used to control the speech of the participant according to the operation instruction.

[0036] In order to solve the above problems, the present invention further discloses an information processing device, which is applied to an AI server and includes:

[0037] A receiving module, used for receiving a video stream of a conference participant sent by the first server;

[0038] An identification module, used for sending the video stream to the AI ​​virtual terminal, so that the AI ​​virtual terminal identifies the video stream and obtains the picture of the participant;

[0039] A detection module, used to detect the participant's picture, and generate an operation instruction when a specified action is detected in the participant's picture;

[0040] The second control module is used to send the operation instruction to the first server, so that the first server controls the speech of the participant according to the operation instruction.

[0041] In order to solve the above problems, the present invention also discloses an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the information processing method when executing the computer program.

[0042] In order to solve the above problem, the present invention further discloses a computer-readable storage medium, which stores a computer program for implementing the information processing method when executed by a processor.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] The present invention pulls in a target AI virtual terminal in a video conference, identifies the video stream of the participant through the target AI virtual terminal to obtain the participant's picture, and the AI ​​server detects the participant's picture, and generates an operation instruction when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction, and the entire control of the participant's speech is achieved through AI technology, that is, the video stream of the participant is identified through the target AI virtual terminal to obtain the participant's picture, and the AI ​​server detects the participant's picture, and then performs related operations. In this way, the operation of the video conference can be achieved without the help of the chairman's operation. Therefore, even if the chairman's terminal fails, the speaker can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0045] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of an information processing method described in an embodiment of the present invention;

[0047] Figure 2 is a flow chart of an information processing method described in an embodiment of the present invention;

[0048] Figure 3 is a flow chart of an information processing method described in an embodiment of the present invention;

[0049] Figure 4 is a structural diagram of an information processing device according to an embodiment of the present invention;

[0050] Figure 5 It is a structural diagram of an information processing device described in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Reference Figure 1 , which shows a flow chart of an information processing method according to an embodiment of the present invention, the method is applied to a first server, and specifically includes:

[0053] Step 101: Pull the target AI virtual terminal into the video conference.

[0054] In actual applications, when the first server completes the establishment of a meeting, it first determines the conference mode of the established meeting. If the conference mode of the video conference belongs to a microphone-grabbing conference, the target AI virtual terminal is pulled into the microphone-grabbing conference.

[0055] Step 102: Send the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant.

[0056] The first server sends the video stream of the participant to the AI ​​virtual terminal through the AI ​​server, and the AI ​​virtual terminal recognizes the video stream to obtain the picture of the participant.

[0057] In practical applications, AI virtual terminals usually use video processing and image processing technologies to convert each frame in the video into an image file. Specifically, Python and OpenCV libraries can be used to implement this. The specific operations are as follows:

[0058] 1. Import library: First import the OpenCV library.

[0059] 2. Define function: Define a function video_to_frames, which accepts the video path and output folder path as parameters.

[0060] 3. Open the video file: Use cv2.VideoCapture to open the video file.

[0061] 4. Read video frames: Use a loop to read each frame of the video.

[0062] 5. Save frame as picture: Save each frame as a picture file with the file name format of frame_XXXX.jpg, where XXXX is the frame number.

[0063] 6. Release the video object: After reading all frames, release the video object.

[0064] Through the above operations, the AI ​​virtual terminal converts the video stream into pictures of the participants.

[0065] Step 103: Receive the operation instruction returned by the AI ​​server.

[0066] Among them, the operation instruction is generated by the AI ​​server when detecting the picture of the participant and detecting the existence of a specified action in the picture of the participant. The specified action includes: an action identifier, which represents the execution of different operations. For example, if the action identifier is raising your hand, it means going on the microphone; if the action identifier is making a cross with both hands, it means leaving the microphone. Other action identifiers can also be used in actual applications, and the present invention does not make any specific restrictions on this.

[0067] In practical applications, AI servers usually rely on computer vision technology and deep learning models to identify the specified actions in the participant's pictures. The following steps are mainly used to implement the specified action recognition in the participant's pictures through AI servers:

[0068] 1. Data collection and preprocessing

[0069] Collect data: First, we need a large number of image datasets with action labels. These datasets usually contain images of various actions.

[0070] Preprocessing: Preprocess the images, including scaling, cropping, normalization, etc., to facilitate better model learning.

[0071] 2. Model selection and training

[0072] Model selection: Commonly used models include convolutional neural networks (CNN), recurrent neural networks (RNN), long short-term memory networks (LSTM), etc. In recent years, Transformer-based models (such as ViT and DETR) have also performed well in image recognition tasks.

[0073] Training the model: Use the labeled dataset to train the model. During the training process, the model will learn how to extract features from images and recognize different actions.

[0074] 3. Feature extraction

[0075] Convolutional layer: The CNN model extracts spatial features from images through multiple convolutional layers.

[0076] Attention Mechanism: Some advanced models such as Transformer use attention mechanisms to focus on key areas in the image.

[0077] 4. Action Classification

[0078] Fully connected layer: The extracted features are classified through the fully connected layer, and the possible action categories and their probabilities in the image are output.

[0079] 5. Post-processing

[0080] Confidence threshold: Filter out low-confidence prediction results based on the confidence threshold.

[0081] Non-maximum suppression (NMS): When detecting multiple actions, NMS is used to remove overlapping detection boxes.

[0082] 6. Deployment and Optimization

[0083] Deployment: Deploy the trained model to the AI ​​server, usually using an inference engine such as TensorFlowServing, ONNX Runtime, etc.

[0084] Step 104: Control the speech of the conference participant according to the operation instruction.

[0085] The first server controls the speech of the participant according to the operation instruction sent by the AI ​​server. If the operation instruction is to switch the speaker, the terminal to which the participant belongs is controlled to speak. If the operation instruction is to leave the microphone, the first server cancels the speaker role of the meeting in real time.

[0086] In this embodiment, a target AI virtual terminal is pulled into the video conference, and the target AI virtual terminal is used to identify the video stream of the participant to obtain the participant's picture, and the AI ​​server detects the participant's picture, and generates an operation instruction when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction, and the entire control of the participant's speech is achieved through AI technology, that is, the target AI virtual terminal is used to identify the participant's video stream to obtain the participant's picture, and the AI ​​server detects the participant's picture and then performs related operations. In this way, the operation of the video conference can be achieved without the help of the chairman's operation. Therefore, even if the chairman's terminal fails, the speaker can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0087] Reference Figure 2 , which shows a flow chart of an information processing method according to an embodiment of the present invention, the method is applied to a first server, and specifically includes:

[0088] In practical applications, the first server may be a conference management server or an autonomous server, and the present invention does not impose any specific limitation on this.

[0089] Step 201: Receive an AI microphone grabbing conference request triggered by a client.

[0090] The client is installed with conference scheduling software (PAMIR), through which the conference is managed.

[0091] In actual applications, an AI microphone grabbing switch is set in the client's conference scheduling software. When the AI ​​microphone grabbing switch is turned on, it means that the conference mode of the video conference is an AI microphone grabbing conference. The client triggers the AI ​​microphone grabbing conference request and sends the AI ​​microphone grabbing conference request to the conference management server.

[0092] Step 202: The first server selects a target AI virtual terminal from the AI ​​virtual terminals of the connected AI servers in response to the AI ​​microphone grabbing conference request.

[0093] In actual applications, the AI ​​server will be configured with multiple AI virtual terminals, and each conference corresponds to one AI virtual terminal.

[0094] Among them, the AI ​​microphone-grabbing conference request carries a video conference identifier (conference identifier) ​​and the main message number of the AI ​​server. The first server responds to the AI ​​microphone-grabbing conference request, selects an AI virtual terminal with idle status information from the AI ​​virtual terminals of the connected AI server according to the video conference identifier and the main message number, and uses the idle AI virtual terminal as the target AI virtual terminal.

[0095] Step 203: Send a response to the client indicating that the conference is successfully started.

[0096] When the client receives a response from the first server indicating that the video conference has been successfully started, it indicates that the video conference is in a microphone-grabbing conference. At this time, the target AI virtual terminal can be pulled into the video conference. The maximum number of participating terminals in the video conference can be 20 or more, and the present invention does not impose any specific limitation on this.

[0097] Since the maximum number of participating terminals in a video conference can be 20, the target AI virtual terminal can also decode 20 video streams simultaneously.

[0098] Step 204: Pull the target AI virtual terminal into the video conference.

[0099] Step 205: Send the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant.

[0100] When the first server receives a response that the target AI virtual terminal has successfully joined the meeting, it sends the video stream of the participant to the target AI virtual terminal through the AI ​​server. The first server saves the correspondence between the participant and the AI ​​stream switching channel number, that is, saves the correspondence between the participant number and the AI ​​virtual terminal.

[0101] The target AI virtual terminal identifies the video stream to obtain the picture of the participant, and sends the picture of the participant to the AI ​​server.

[0102] Step 206: Receive the operation instruction returned by the AI ​​server.

[0103] The AI ​​server detects the participant's picture and generates an operation instruction when it detects that the participant's picture contains the specified action of the participant's terminal. In actual applications, there may be specified actions of multiple terminals in the participant's picture. After the operation instruction is generated, different participants can be controlled to speak according to the priority of the terminal.

[0104] Step 207: Control the speech of the conference participant according to the operation instruction.

[0105] The first server parses the operation instruction to obtain a specified action identifier and a channel number; determines the participant from the corresponding relationship between the participant and the AI ​​switching channel number according to the channel number; and controls the speech of the participant according to the specified action identifier.

[0106] For example, the specified action identifier obtained from the operation instruction is microphone, the channel number is 1, the AI ​​stream switching channel number is 0-19, the participant is 0-19, and the AI ​​stream switching channel number corresponds to the participant one-to-one. Then, according to channel number 1, from the correspondence between the participant and the AI ​​stream switching channel number, it is determined that the participant is 1, and the participant 1 is controlled to speak.

[0107] In actual applications, when the first server receives an AI microphone grabbing conference request to close a video conference, the first server removes the target AI virtual terminal from the video conference and notifies the AI ​​server to unbind the target AI virtual terminal. Through the above operations, the speaker in the video conference can be removed in real time, reducing the pressure on the first server.

[0108] In this embodiment, a target AI virtual terminal is pulled into the video conference, and the target AI virtual terminal is used to identify the video stream of the participant to obtain the participant's picture, and the AI ​​server detects the participant's picture, and generates an operation instruction when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction, and the entire control of the participant's speech is achieved through AI technology, that is, the target AI virtual terminal is used to identify the participant's video stream to obtain the participant's picture, and the AI ​​server detects the participant's picture and then performs related operations. In this way, the operation of the video conference can be achieved without the help of the chairman's operation. Therefore, even if the chairman's terminal fails, the speaker can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0109] Reference Figure 3 , which shows a flow chart of an information processing method according to an embodiment of the present invention, the method is applied to an AI server, and specifically includes:

[0110] Step 301: Receive a video stream of a conference participant sent by a first server.

[0111] In actual applications, before receiving the video stream of the participant sent by the first server, the method also includes: configuring an AI virtual terminal on the AI ​​server; when receiving an AI microphone grabbing conference request triggered by the client, selecting a target AI virtual terminal from the configured AI virtual terminals.

[0112] Step 302: Send the video stream to the AI ​​virtual terminal so that the AI ​​virtual terminal recognizes the video stream and obtains the picture of the participant.

[0113] Step 303: Detect the participant's picture, and generate an operation instruction when a specified action is detected in the participant's picture.

[0114] Step 304: Send the operation instruction to the first server, so that the first server controls the speech of the participant according to the operation instruction.

[0115] In this embodiment, a target AI virtual terminal is pulled into the video conference, and the target AI virtual terminal is used to identify the video stream of the participant to obtain the participant's picture, and the AI ​​server detects the participant's picture, and generates an operation instruction when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction, and the entire control of the participant's speech is achieved through AI technology, that is, the target AI virtual terminal is used to identify the participant's video stream to obtain the participant's picture, and the AI ​​server detects the participant's picture and then performs related operations. In this way, the operation of the video conference can be achieved without the help of the chairman's operation. Therefore, even if the chairman's terminal fails, the speaker can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0116] It should be noted that, for the above method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the present invention.

[0117] Based on the description of the above method embodiments, the present invention also provides corresponding device embodiments to implement the contents described in the above method embodiments.

[0118] Reference Figure 4 , which shows a structural diagram of an information processing device according to an embodiment of the present invention, the device is applied to a first server, and specifically includes:

[0119] A conference module 401 is used to connect the target AI virtual terminal to a video conference;

[0120] The sending module 402 is used to send the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the AI ​​virtual terminal can identify the video stream and obtain the picture of the participant;

[0121] The return module 403 is used to receive the operation instruction returned by the AI ​​server, where the operation instruction is generated when the AI ​​server detects the participant's picture and detects that a specified action exists in the participant's picture;

[0122] The first control module 404 is used to control the speech of the participant according to the operation instruction.

[0123] Optionally, the device further comprises:

[0124] The first receiving module is used to receive an AI microphone grabbing conference request triggered by a client;

[0125] A first selection module, configured for the first server to select a target AI virtual terminal from AI virtual terminals of connected AI servers in response to the AI ​​microphone grabbing conference request;

[0126] The opening module is used to send a response to the client indicating that the microphone grabbing conference has been successfully opened.

[0127] Optionally, the AI ​​microphone grabbing conference request carries a conference identifier and a main message number of the AI ​​server;

[0128] The first selection module is specifically used for the first server to respond to the AI ​​microphone grabbing conference request, select an AI virtual terminal with idle status information from the AI ​​virtual terminals of the connected AI server according to the conference identifier and the main message number, and use the idle AI virtual terminal as the target AI virtual terminal.

[0129] Optionally, the first control module includes:

[0130] A parsing unit, used to parse the operation instruction to obtain a specified action identifier and a channel number;

[0131] a determination unit, configured to determine the conference participant from the correspondence between the conference participant and the AI ​​stream switching channel number according to the channel number;

[0132] The control unit is used to control the speech of the conference participant according to the specified action identifier.

[0133] Optionally, it is characterized in that the device further comprises:

[0134] The removal module is used to, when receiving an AI microphone grabbing conference request to close the video conference, the first server removes the target AI virtual terminal from the video conference and notifies the AI ​​server to unbind the target AI virtual terminal.

[0135] In this embodiment, a target AI virtual terminal is pulled into the video conference, and the target AI virtual terminal is used to identify the video stream of the participant to obtain the participant's picture. The AI ​​server detects the participant's picture and generates it when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction. The entire control of the participant's speech is achieved through AI technology, and does not require the operation of the chairman. In this way, even if the terminal belonging to the chairman fails, the speaker's operation can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0136] Reference Figure 5 , which shows a structural diagram of an information processing device according to an embodiment of the present invention, the device is applied to an AI server, and specifically includes:

[0137] The second receiving module 501 is used to receive the video stream of the conference participant sent by the first server;

[0138] The recognition module 502 is used to send the video stream to the AI ​​virtual terminal so that the AI ​​virtual terminal recognizes the video stream and obtains the picture of the participant;

[0139] The detection module 503 is used to detect the participant picture and generate an operation instruction when a specified action is detected in the participant picture;

[0140] The second control module 504 is used to send the operation instruction to the first server, so that the first server controls the speech of the participant according to the operation instruction.

[0141] Optionally, the device further comprises:

[0142] A configuration module, used for configuring an AI virtual terminal on the AI ​​server;

[0143] The second selection module is used to select a target AI virtual terminal from the configured AI virtual terminals when receiving an AI microphone grabbing conference request triggered by a client.

[0144] In this embodiment, a target AI virtual terminal is pulled into the video conference, and the target AI virtual terminal is used to identify the video stream of the participant to obtain the participant's picture. The AI ​​server detects the participant's picture and generates it when a specified action is detected in the participant's picture; the speech of the participant is controlled according to the operation instruction. The entire control of the participant's speech is achieved through AI technology, and does not require the operation of the chairman. In this way, even if the terminal belonging to the chairman fails, the speaker's operation can still be switched in real time, thereby ensuring the smooth progress of the video conference.

[0145] As for the above-mentioned device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts may refer to the partial description of the method embodiment.

[0146] An embodiment of the present invention further provides an electronic device, including:

[0147] A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the information processing method is implemented when the processor executes the computer program.

[0148] An embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program that implements the information processing method when executed by a processor.

[0149] Alternatively, the computer-readable storage medium may be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0150] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0151] It is easy for a person skilled in the art to think that any combination of the above embodiments is feasible, so any combination of the above embodiments is an implementation scheme of the present invention, but due to space limitations, this specification will not be described in detail here. Although the preferred embodiments of the present invention have been described, once a person skilled in the art knows the basic creative concept, he can make other changes and modifications to these embodiments. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0152] The above is a detailed introduction to an information processing method, device, equipment and medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for a person skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An information processing method, characterized in that: The method is applied to a first server, comprising: Pull the target AI virtual terminal into the video conference; Sending the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant; receiving an operation instruction returned by the AI ​​server, where the operation instruction is generated when the AI ​​server detects the participant's picture and detects that a specified action exists in the participant's picture; The speech of the conference participant is controlled according to the operation instruction.

2. The method according to claim 1, characterized in that Before pulling the target AI virtual terminal into the video conference, the method further includes: Receive AI microphone grabbing conference request triggered by client; The first server selects a target AI virtual terminal from the AI ​​virtual terminals of the connected AI servers in response to the AI ​​microphone grabbing conference request; A response is sent to the client indicating that the conference is successfully started.

3. The method according to claim 2, characterized in that The AI ​​microphone grabbing conference request carries a conference identifier and a main message number of the AI ​​server; The first server selects a target AI virtual terminal from the AI ​​virtual terminals of the connected AI server in response to the AI ​​microphone grabbing conference request, including: The first server responds to the AI ​​microphone grabbing conference request, selects an AI virtual terminal with idle status information from the AI ​​virtual terminals of the connected AI server according to the conference identifier and the main message number, and uses the idle AI virtual terminal as the target AI virtual terminal.

4. The method according to claim 1, characterized in that: The controlling the speech of the participant according to the operation instruction includes: Parsing the operation instruction to obtain a specified action identifier and a channel number; Determine the participant from the corresponding relationship between the participant and the AI ​​switching channel number according to the channel number; The speech of the conference participant is controlled according to the designated action identifier.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: When receiving the AI ​​microphone grabbing conference request to close the video conference, the first server removes the target AI virtual terminal from the video conference and notifies the AI ​​server to unbind the target AI virtual terminal.

6. An information processing method, characterized in that: The method is applied to an AI server and includes: Receiving a video stream of a conference participant sent by the first server; Sending the video stream to the AI ​​virtual terminal so that the AI ​​virtual terminal recognizes the video stream and obtains the picture of the participant; Detecting the participant's picture, and generating an operation instruction when a specified action is detected in the participant's picture; The operation instruction is sent to the first server, so that the first server controls the speech of the participant according to the operation instruction.

7. The method according to claim 6, characterized in that Before receiving the video stream of the conference participant sent by the first server, the method further includes: Configuring an AI virtual terminal on the AI ​​server; When receiving an AI microphone grabbing conference request triggered by a client, a target AI virtual terminal is selected from the configured AI virtual terminals.

8. An information processing device, characterized in that: The device is applied to a first server and includes: The conference module is used to connect the target AI virtual terminal to the video conference; A sending module, used to send the video stream of the participant to the target AI virtual terminal through the AI ​​server, so that the target AI virtual terminal recognizes the video stream and obtains the picture of the participant; A return module, configured to receive an operation instruction returned by the AI ​​server, wherein the operation instruction is generated when the AI ​​server detects the participant's picture and detects that a specified action exists in the participant's picture; The first control module is used to control the speech of the participant according to the operation instruction.

9. An information processing device, characterized in that: The device is applied to an AI server, comprising: A receiving module, used for receiving a video stream of a conference participant sent by the first server; An identification module, used for sending the video stream to the AI ​​virtual terminal, so that the AI ​​virtual terminal identifies the video stream and obtains the picture of the participant; A detection module, used to detect the participant's picture, and generate an operation instruction when a specified action is detected in the participant's picture; The second control module is used to send the operation instruction to the first server, so that the first server controls the speech of the participant according to the operation instruction.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the information processing method according to any one of claims 1 to 5 or claims 6 to 7 is implemented.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program that, when executed by a processor, implements the information processing method according to any one of claims 1 to 5 or claims 6 to 7.

Images