Subtitle display method and system for video conference, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a subtitle display method and system for a video conference, and is applied to a server, and the method comprises the steps: obtaining a real-time subtitle opening message of the video conference; applying for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle opening message; transmitting an audio stream of the video conference to an artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to an artificial intelligence server, and receiving a subtitle result of subtitle translation according to the voice recognition interface from the artificial intelligence server; and outputting the subtitle result to a participating terminal of the video conference, so that the participating terminal displays the subtitle result. According to the embodiment of the invention, the accuracy, the real-time performance and the adaptability of the subtitles are improved by combining the artificial intelligence technology with the subtitle terminal, the information transmission effect is greatly improved especially under the scene of multi-party session or special user requirements (such as hearing impairment), and the participating experience is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of video conferencing, and in particular to a subtitle display method for a video conference, a subtitle display system for a video conference, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the growing demand for remote work and online learning, video conferencing has become a common tool for modern work and learning. Whether it is a company, school or individual, video conferencing can achieve efficient communication and collaboration without geographical restrictions.

[0003] However, there is often a problem in video conferencing: the speaker's content cannot be effectively transmitted to all participants. Especially in multi-person conversations or when there are participants with hearing impairments, clear information transmission is particularly important. Summary of the invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a method for displaying subtitles in a video conference, a system for displaying subtitles in a video conference, an electronic device, and a computer-readable storage medium that overcome the above problems or at least partially solve the above problems.

[0005] In order to solve the above problems, an embodiment of the present invention discloses a subtitle display method for a video conference, which is applied to a server. The method includes: obtaining a real-time subtitle start message of the video conference; applying for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle start message; transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives subtitle results from the artificial intelligence server that are translated according to a speech recognition interface; outputting the subtitle results to the participant terminals of the video conference, so that the participant terminals display the subtitle results.

[0006] Optionally, applying for an artificial intelligence subtitle terminal from the artificial intelligence server according to the real-time subtitle start message includes: reading conference identification information of the video conference from the real-time subtitle start message; transmitting the conference identification information to the artificial intelligence server, so that the artificial intelligence server allocates the artificial intelligence subtitle terminal to the video conference according to the conference identification information.

[0007] Optionally, before transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal, the method also includes: assigning a subtitle terminal role to the artificial intelligence subtitle terminal, and adding the artificial intelligence subtitle terminal to the video conference.

[0008] Optionally, transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal includes: obtaining a subtitle switch status message from the client; determining the audio stream based on the subtitle switch status message and the current speaker of the video conference; and transmitting the audio stream to the artificial intelligence subtitle terminal.

[0009] Optionally, determining the audio stream based on the subtitle switch status message and the current speaker of the video conference includes: identifying identification information of a speaker in a subtitle-on state from the subtitle switch status message; determining whether there is a speaker corresponding to the speaker identification information among the current speakers of the video conference; and obtaining the audio stream if there is a speaker corresponding to the speaker identification information among the current speakers of the video conference.

[0010] Optionally, outputting the subtitle results to the participant terminal of the video conference so that the participant terminal displays the subtitle results, includes: transmitting the subtitle results in a JSON structure to the artificial intelligence subtitle terminal; transmitting the subtitle results to the participant terminal through an output channel of the artificial intelligence subtitle terminal, so that the participant terminal parses the subtitle results in the JSON structure to obtain subtitle content, and displays the subtitle content according to the speaker corresponding to the subtitle results.

[0011] Optionally, after outputting the subtitle results to the participating terminal of the video conference so that the participating terminal displays the subtitle results, the method also includes: obtaining a real-time subtitle closing message of the video conference; removing the artificial intelligence subtitle terminal from the video conference according to the real-time subtitle closing message, and notifying the artificial intelligence server to release the artificial intelligence subtitle terminal.

[0012] An embodiment of the present invention also discloses a subtitle display system for a video conference, which is applied to a server, and the system includes: a subtitle start message acquisition module, which is used to obtain a real-time subtitle start message for a video conference; a subtitle terminal application module, which is used to apply for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle start message; an audio stream transmission module, which is used to transmit the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives subtitle results from the artificial intelligence server that are translated according to a speech recognition interface; a subtitle result transmission module, which is used to output the subtitle results to the participant terminals of the video conference, so that the participant terminals display the subtitle results.

[0013] Optionally, the subtitle terminal application module includes: a conference identification reading module, used to read the conference identification information of the video conference from the real-time subtitle start message; a conference identification transmission module, used to transmit the conference identification information to the artificial intelligence server, so that the artificial intelligence server allocates the artificial intelligence subtitle terminal to the video conference according to the conference identification information.

[0014] Optionally, the system also includes: a subtitle terminal joining module, which is used to assign a subtitle terminal role to the artificial intelligence subtitle terminal and add the artificial intelligence subtitle terminal to the video conference before the audio stream transmission module transmits the audio stream of the video conference to the artificial intelligence subtitle terminal.

[0015] Optionally, the audio stream transmission module includes: a switch status acquisition module, used to obtain a subtitle switch status message from the client; an audio stream determination module, used to determine the audio stream based on the subtitle switch status message and the current speaker of the video conference; and an audio stream sending module, used to transmit the audio stream to the artificial intelligence subtitle terminal.

[0016] Optionally, the audio stream determination module includes: a speaker identification recognition module, used to identify the speaker identification information of the speaker in the subtitle on state from the subtitle switch status message; a speaker judgment module, used to judge whether there is a speaker corresponding to the speaker identification information among the current speakers of the video conference; an audio stream acquisition module, used to acquire the audio stream if there is a speaker corresponding to the speaker identification information among the current speakers of the video conference.

[0017] Optionally, the subtitle result transmission module includes: a result sending module, used to transmit the subtitle result of the JSON structure to the artificial intelligence subtitle terminal; a subtitle output module, used to transmit the subtitle result to the participant terminal through the output channel of the artificial intelligence subtitle terminal, so that the participant terminal parses the subtitle result of the JSON structure to obtain the subtitle content, and displays the subtitle content according to the speaker corresponding to the subtitle result.

[0018] Optionally, the system also includes: a closing message acquisition module, used to obtain a real-time subtitle closing message of the video conference after the subtitle result transmission module outputs the subtitle result to the participating terminal of the video conference so that the participating terminal displays the subtitle result; a subtitle terminal removal and release module, used to remove the artificial intelligence subtitle terminal from the video conference according to the real-time subtitle closing message, and notify the artificial intelligence server to release the artificial intelligence subtitle terminal.

[0019] An embodiment of the present invention also discloses an electronic device, comprising: one or more processors; and one or more machine-readable media storing instructions thereon, which, when executed by the one or more processors, enables the electronic device to perform the subtitle display method for video conferencing as described above.

[0020] The embodiment of the present invention further discloses a computer-readable storage medium, which stores a computer program that enables a processor to execute the above-mentioned subtitle display method for a video conference.

[0021] The embodiments of the present invention include the following advantages:

[0022] The subtitle display scheme for video conferencing provided by an embodiment of the present invention is applied to a server. The server obtains a real-time subtitle start message of the video conference, and then applies to an artificial intelligence (AI) server for an artificial intelligence subtitle terminal according to the real-time subtitle start message. Next, the audio stream of the video conference is transmitted to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle result of subtitle translation according to the speech recognition interface from the artificial intelligence server. Finally, the subtitle result is output to the participant terminal of the video conference, so that the participant terminal displays the subtitle result.

[0023] The embodiment of the present invention obtains the real-time subtitle opening message of the video conference through the server, and processes the audio stream with the help of the artificial intelligence server and the subtitle terminal, so as to generate subtitles in real time and transmit them to the participating terminals. The audio stream of the speaker is identified and translated by artificial intelligence technology, which solves the problem that the content of the speaker cannot be effectively transmitted to all participants in the background technology. Especially for multi-person conversation scenarios, subtitles can more accurately correspond to the speaker's content, improving the efficiency and accuracy of information transmission. By introducing the voice recognition interface of the artificial intelligence server, the audio stream is subtitled. Compared with the traditional subtitle generation method, artificial intelligence technology can better adapt to complex voice environments, improve recognition accuracy, and reduce the impact of accent differences, excessive speech speed, etc. on subtitle quality. It helps to reduce the error rate of subtitles, make subtitle results more accurate, and improve the understanding experience of participants. It ensures that the participating terminal displays real-time subtitle results, thereby providing a more friendly video conference experience for the hearing-impaired. Participants with hearing impairments can follow up the content of the speaker in real time through accurate subtitle information, solving the problem of poor information transmission in the background technology. It can meet the needs of different types of participants, especially improve the barrier-free participation of video conferences. It can process audio streams and transmit them to the artificial intelligence server for recognition, so it has the ability to handle speeches by multiple people. Through precise speech recognition and subtitle translation, the content of each speaker can be accurately displayed, which solves the problem that the speech content cannot be effectively distinguished and transmitted during multi-person conversations in the background technology, and improves the clarity of the conversation. The complex computing tasks such as speech recognition and subtitle translation are handed over to the artificial intelligence server and artificial intelligence subtitle terminal for processing, and the participating terminals only need to display the subtitle results. The burden on the participating terminals is greatly reduced, and there is no need to undertake complex speech processing tasks, which saves hardware resources and ensures a smooth conference experience for participants, especially when the device performance is limited. High-quality subtitle services can still be obtained. Allow participants to dynamically enable subtitle services as needed. This flexibility meets the needs of different conference scenarios. When subtitles are not needed, the function can be turned off, saving resources and improving the intelligence and personalized experience of subtitle display.

[0024] In summary, the embodiments of the present invention improve the accuracy, real-time and adaptability of subtitles by combining artificial intelligence technology and subtitle terminals, especially in scenarios of multi-party conversations or special user needs (such as hearing impairment), which greatly improves the effect of information transmission and enhances the conference experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flowchart of the steps of a subtitle display method for a video conference according to an embodiment of the present invention;

[0026] Figure 2This is an interactive schematic diagram of a solution for displaying real-time subtitles of a speaker through AI voice recognition based on a visual network according to an embodiment of the present invention;

[0027] Figure 3 The present invention is a structural block diagram of a subtitle display system for a video conference according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] 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.

[0029] The subtitle display scheme for video conferencing proposed in the embodiment of the present invention aims to achieve efficient real-time subtitle display by introducing artificial intelligence technology. In the scheme, the server first obtains the real-time subtitle start message of the video conference, then applies for the subtitle terminal from the artificial intelligence server, receives the audio stream and transmits it to the subtitle terminal. The subtitle terminal transmits the audio stream to the artificial intelligence server, generates subtitle results through the voice recognition interface, and feeds these results back to the participating terminals to ensure the accurate display of subtitles. This scheme not only improves the accuracy and timeliness of subtitle generation, but also effectively supports the subtitle display of multi-person conversations, especially providing a convenient solution for the hearing-impaired and cross-language conferences, while reducing the processing burden of the participating terminals.

[0030] Reference Figure 1 , shows a flowchart of the steps of a subtitle display method for a video conference according to an embodiment of the present invention. The subtitle display method for a video conference is applied to a server, and the subtitle display method for a video conference may specifically include the following steps:

[0031] Step 101: Obtain a real-time subtitle activation message for a video conference.

[0032] In an embodiment of the present invention, when a video conference starts, the server does not start the subtitle display function by default, and the subtitle display will only be triggered under certain specific conditions. The main function of this step is to ensure that the subtitle function is only started when needed to save resources and meet user needs. The real-time subtitle activation message can be triggered in a variety of ways. Generally, the conditions for subtitle activation include a manual request from the user, the video conferencing system automatically detecting the participation of the hearing-impaired, or the conference host requesting to activate the subtitle service. For example, on some video conferencing platforms, participants can request to activate the subtitles by clicking a specific button, and this operation will trigger the system to generate a "real-time subtitle activation message". This message is an instruction that notifies the server to start the subtitle service and make corresponding resource allocations.

[0033] For example, in a company meeting, participants come from different countries and regions, and some of them may have hearing impairments. After the meeting starts, the host finds that some participants need subtitle assistance, so he manually clicks the "Turn on subtitles" button. After receiving the instruction, the conference system generates a "real-time subtitle turn-on message" to inform the server that the subtitle service can be started. This process ensures that the subtitle display function is started based on actual needs, thereby improving the utilization efficiency of system resources and avoiding unnecessary burdens.

[0034] In addition to manual triggering, some advanced systems can automatically detect the need based on the participant's history or profile. For example, the conference system can identify the preferences or obstacles of certain users and automatically send a subtitle-enabling message when they enter the meeting. In this case, the entire subtitle display function can be more intelligent and personalized, ensuring that every user can get the best meeting experience.

[0035] Step 102: Apply for an artificial intelligence subtitle terminal from the artificial intelligence server according to the real-time subtitle activation message.

[0036] In an embodiment of the present invention, the artificial intelligence subtitle terminal is the core component of the entire subtitle generation process, responsible for processing the audio stream, and finally generating the subtitle results by collaborating with the artificial intelligence server. In this step, the server acts as a command center to coordinate the resource allocation of the subtitle service and ensure the seamless connection between the subtitle terminal and the artificial intelligence server. When the server receives the subtitle start message, it needs to contact the artificial intelligence server to apply for an artificial intelligence subtitle terminal specifically for processing the meeting. The artificial intelligence subtitle terminal is an intelligent device or virtual instance that has voice processing, voice recognition and other capabilities, can receive audio streams in real time and interact with the artificial intelligence server. The step of applying for a subtitle terminal ensures that the subtitle service can be carried out smoothly. Specifically, the server will reasonably allocate artificial intelligence resources based on factors such as the scale of the meeting, the number of participants, and the duration of the meeting.

[0037] For example, at an academic seminar, the host requests that the subtitle service be turned on. After receiving the instruction, the server will quickly communicate with the AI ​​server and request a subtitle terminal. Since the conference involves researchers from multiple countries, the server may choose to configure a high-performance AI subtitle terminal that can handle multi-language input and subtitle generation for long meetings. This process ensures that the subtitle terminal can operate normally and meet the requirements of the specific conference.

[0038] In addition, this step also involves the dynamic allocation of AI resources. The needs of different meetings may vary, and the AI ​​server needs to intelligently allocate appropriate subtitle terminals. For example, some small meetings may only require basic subtitle services, while large international meetings may require higher processing power. In this process, the server dynamically allocates appropriate subtitle terminals for each meeting by collaborating with the AI ​​server to ensure the efficient operation of the subtitle service.

[0039] Step 103, transmit the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle results of subtitle translation based on the speech recognition interface from the artificial intelligence server.

[0040] In an embodiment of the present invention, the audio stream is transmitted to the artificial intelligence subtitle terminal through the network structure of the conference system, and then uploaded to the artificial intelligence server for processing through the subtitle terminal. After receiving these audio streams, the artificial intelligence server translates the audio streams with the help of the speech recognition interface to generate corresponding subtitle results. The subtitle terminal then receives these subtitle results and prepares to transmit them back to each participating terminal of the conference.

[0041] In actual application, when the subtitle terminal is started and ready, the conference system will capture the audio stream in real time. These audio streams contain voice content, and the conference system compresses and packages the audio streams through encoding technology for more efficient transmission. At this time, the audio stream is sent to the artificial intelligence subtitle terminal, which performs preliminary processing on the audio stream, such as eliminating background noise and enhancing audio quality. After completing these preprocessing, the audio stream will be further uploaded to the artificial intelligence server for speech recognition processing.

[0042] The AI ​​server uses advanced speech recognition algorithms to analyze and process the audio stream and generate text content that matches the speech content. The speech recognition technology here can apply natural language processing and deep learning models to improve the accuracy of recognition. Challenges in this process include dealing with different accents, speaking speeds, intonations, and distinguishing the speech streams of different speakers in a multi-person conversation.

[0043] For example, in a corporate strategy meeting, multiple management personnel take turns to speak. Everyone's pronunciation habits, speaking speed, and accent are different, and these audio streams are captured in real time by the conference system and sent to the AI ​​subtitle terminal. The subtitle terminal uploads the audio stream to the AI ​​server, which uses speech recognition technology to translate the audio stream into subtitle results. This process ensures that high-quality and accurate subtitles can be generated regardless of the speaker's language or accent.

[0044] Step 104: output the subtitle result to the participant terminal of the video conference so that the participant terminal displays the subtitle result.

[0045] In the embodiment of the present invention, the display of subtitles needs to consider multiple factors, including screen layout, font size, color contrast, etc. The subtitles should be as clear and easy to read as possible, and synchronized with the speech content so that participants can follow the speaker's speech content in real time. Participants can also customize the display of subtitles according to personal needs, such as adjusting the position of subtitles, switching subtitles in different languages, etc.

[0046] For example, in an academic seminar, participants come from many countries, some of whom speak in their native language, while others rely on subtitles to understand the content. When the subtitle results are transmitted to the participating terminals through the server, the participating terminals will display the corresponding language subtitles according to the settings. For content that needs to be translated, the subtitles will be displayed in the appropriate language, and for users who do not need translation, the subtitles will directly display the original language of the speaker. This process ensures that each participant can get a personalized conference experience while ensuring the clarity and accuracy of the subtitle display.

[0047] The subtitle display scheme for video conferencing provided by an embodiment of the present invention is applied to a server. The server obtains a real-time subtitle start message of the video conference, and then applies to an artificial intelligence server for an artificial intelligence subtitle terminal according to the real-time subtitle start message. Next, the audio stream of the video conference is transmitted to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle result of subtitle translation according to the speech recognition interface from the artificial intelligence server. Finally, the subtitle result is output to the participant terminal of the video conference, so that the participant terminal displays the subtitle result.

[0048] The embodiment of the present invention obtains the real-time subtitle opening message of the video conference through the server, and processes the audio stream with the help of the artificial intelligence server and the subtitle terminal, so as to generate subtitles in real time and transmit them to the participating terminals. The audio stream of the speaker is identified and translated by artificial intelligence technology, which solves the problem that the content of the speaker cannot be effectively transmitted to all participants in the background technology. Especially for multi-person conversation scenarios, subtitles can more accurately correspond to the speaker's content, improving the efficiency and accuracy of information transmission. By introducing the voice recognition interface of the artificial intelligence server, the audio stream is subtitled. Compared with the traditional subtitle generation method, artificial intelligence technology can better adapt to complex voice environments, improve recognition accuracy, and reduce the impact of accent differences, excessive speech speed, etc. on subtitle quality. It helps to reduce the error rate of subtitles, make subtitle results more accurate, and improve the understanding experience of participants. It ensures that the participating terminal displays real-time subtitle results, thereby providing a more friendly video conference experience for the hearing-impaired. Participants with hearing impairments can follow up the content of the speaker in real time through accurate subtitle information, solving the problem of poor information transmission in the background technology. It can meet the needs of different types of participants, especially improve the barrier-free participation of video conferences. It can process audio streams and transmit them to the artificial intelligence server for recognition, so it has the ability to handle speeches by multiple people. Through precise speech recognition and subtitle translation, the content of each speaker can be accurately displayed, which solves the problem that the speech content cannot be effectively distinguished and transmitted during multi-person conversations in the background technology, and improves the clarity of the conversation. The complex computing tasks such as speech recognition and subtitle translation are handed over to the artificial intelligence server and artificial intelligence subtitle terminal for processing, and the participating terminals only need to display the subtitle results. The burden on the participating terminals is greatly reduced, and there is no need to undertake complex speech processing tasks, which saves hardware resources and ensures a smooth conference experience for participants, especially when the device performance is limited. High-quality subtitle services can still be obtained. Allow participants to dynamically enable subtitle services as needed. This flexibility meets the needs of different conference scenarios. When subtitles are not needed, the function can be turned off, saving resources and improving the intelligence and personalized experience of subtitle display.

[0049] In summary, the embodiments of the present invention improve the accuracy, real-time and adaptability of subtitles by combining artificial intelligence technology and subtitle terminals, especially in scenarios of multi-party conversations or special user needs (such as hearing impairment), which greatly improves the effect of information transmission and enhances the conference experience.

[0050] In an exemplary embodiment of the present invention, an implementation method of applying for an artificial intelligence subtitle terminal from an artificial intelligence server according to a real-time subtitle enable message is to read the conference identification information of the video conference from the real-time subtitle enable message; transmit the conference identification information to the artificial intelligence server, so that the artificial intelligence server allocates an artificial intelligence subtitle terminal to the video conference according to the conference identification information. In this implementation, when the server receives a real-time subtitle enable message, it is necessary to read the conference identification information of the video conference from the message. The conference identification information is a unique identifier used to distinguish different video conferences. Through the conference identification information, the specific instance of the current meeting can be identified, thereby providing customized services for each meeting, including subtitle display functions.

[0051] Specifically, when a subtitle activation message is triggered, the server retrieves the content of the message and reads out the conference identification information. The conference identification information may include information such as conference ID, organizer information, and conference time. This information is used to uniquely identify the current video conference to ensure that the server does not mistakenly allocate resources to other conferences. Subsequently, the conference identification information is transmitted to the artificial intelligence server, which allocates an artificial intelligence subtitle terminal to the specific conference based on the conference identification information. This subtitle terminal is dedicated to the current video conference.

[0052] For example, suppose a company is holding a meeting with the meeting number "Global2024". When the meeting starts, the host decides to enable the real-time subtitle function. The conference system generates a real-time subtitle activation message, which contains the conference identifier "Global2024". After the server reads this identifier, it sends it to the AI ​​server. The AI ​​server automatically assigns a dedicated AI subtitle terminal to the meeting based on the identifier "Global2024". This subtitle terminal is responsible for receiving the audio stream of the meeting and sending the audio data to the AI ​​server for processing. After speech recognition, the generated subtitles will be transmitted back to the participants' devices in real time, so that all participants can see the speaker's speech content synchronously.

[0053] In an embodiment of the present invention, the use of conference identification information makes the subtitle processing process of each conference completely independent, and will not be confused due to interference from other conferences. Each conference has its own independent subtitle terminal, which can ensure the customization and pertinence of the subtitle generation process and adapt to the conference requirements of different scales, languages ​​or contents. Using conference identification information to apply for an artificial intelligence subtitle terminal from an artificial intelligence server can also achieve dynamic allocation and optimization of resources. For example, in a small meeting, a relatively basic subtitle terminal can be allocated to the meeting to handle a smaller data flow; while in a large international conference, a higher-performance subtitle terminal can be allocated to cope with complex voice input, multilingual environment or long-term conference requirements. In this way, the subtitle requirements of different meetings can be met to the maximum extent without wasting resources. When there are multiple meetings going on at the same time, independent resources can be accurately allocated to each meeting based on the conference identification information, avoiding resource conflicts or confusion. For example, assuming that a company is holding multiple divisional meetings at the same time, the conference identification information of each meeting is unique. The artificial intelligence server can provide exclusive subtitle terminal services for each meeting based on this information, without worrying about the subtitles of a certain meeting being mistakenly allocated to other meetings.

[0054] In an exemplary embodiment of the present invention, before transmitting the audio stream of a video conference to an artificial intelligence subtitle terminal, an implementation method is to assign a subtitle terminal role to the artificial intelligence subtitle terminal and add the artificial intelligence subtitle terminal to the video conference. In actual application, the artificial intelligence subtitle terminal needs to be assigned a specific role. This role is not only a label to identify its function, but also determines the authority and function of the subtitle terminal in the video conference. The subtitle terminal role means that the terminal will assume responsibilities related to subtitle generation. Explicitly assigning this role in the conference system can ensure that the subtitle terminal has the required permissions to perform its tasks. For example, the subtitle terminal needs to access the audio stream of the conference and understand the speaker's speech content, all of which depend on the specific role and permissions it is granted.

[0055] After assigning roles, you also need to formally add the AI ​​subtitle terminal to the video conference. Adding a subtitle terminal to a conference is similar to adding a new participant. This operation ensures that the subtitle terminal can receive the audio stream in the conference and can keep pace with other participating terminals. The importance of this step is that it makes the subtitle terminal an effective part of the conference and can obtain the speaker's audio stream in real time for processing.

[0056] For example: suppose that in a multi-department video conference, the enterprise wants to provide real-time subtitles for all participants. First, an AI subtitle terminal is assigned the role of "Subtitle Generator" and is granted permission to receive the conference audio stream. Then, the subtitle terminal is added to the conference so that it can receive the audio streams of all speakers in the conference like other participants. After joining the conference, the subtitle terminal automatically starts to monitor the speeches and forwards these audio streams to the AI ​​server for speech recognition and translation.

[0057] By assigning specific roles to artificial intelligence subtitle terminals, the embodiments of the present invention can better manage various types of terminal devices and ensure that each device performs corresponding tasks. This modular role assignment is more flexible when handling multiple tasks. The subtitle terminal focuses on subtitle generation, while other terminals are responsible for video, audio transmission or other conference functions. By explicitly assigning specific roles to the subtitle terminal, the access rights of the subtitle terminal can be controlled to ensure that it can only access data related to subtitle generation. For example, the subtitle terminal does not need to access video streams or other sensitive information, and is only allowed to receive audio streams for translation, which can improve the security and privacy protection of data processing. After adding the subtitle terminal to the meeting, it can obtain the audio stream in real time to ensure the synchronization and accuracy of subtitle generation. Once the subtitle terminal joins the meeting, it can immediately start receiving audio and start the translation process to reduce delays. Participants do not need to perform additional operations or configure the subtitle function. By automatically assigning roles to the subtitle terminal and adding it to the meeting, the seamless integration and automated operation of the entire subtitle service are ensured. The operation steps in the meeting are simplified and the user experience is improved.

[0058] In an exemplary embodiment of the present invention, an implementation method of transmitting the audio stream of a video conference to an artificial intelligence subtitle terminal is to obtain a subtitle switch status message from the client; determine the audio stream according to the subtitle switch status message and the current speaker of the video conference; and transmit the audio stream to the artificial intelligence subtitle terminal. In actual application, the server needs to obtain a subtitle switch status message from the client. The subtitle switch status message is an important instruction that determines whether to turn the subtitle function on or off in the current meeting. This status message is usually triggered by a user (such as a host or conference administrator) manually through the client interface, and can also be automatically generated by the conference system according to preset conditions. For example, in some cases, specific participants in the meeting may manually turn on the subtitle function through the client interface, while in other cases, the conference system may automatically detect that some participants need subtitle support and automatically generate a subtitle switch status message.

[0059] For example, in a multilingual international conference, some participants may need real-time subtitles to understand the speaker's content. At this time, a participant can manually turn on the subtitle function through the client. After the server receives this switch status message, it will decide whether to start the subtitle generation service based on the specific situation.

[0060] Once the server receives the subtitle switch status message, it determines the audio stream that needs to be processed based on the message and the speaker in the current conference. The key step here is to identify the main speaker in the current conference and extract the relevant audio stream from it. The audio stream is the real-time data of the speaker's voice, which is the basis for subsequent subtitle generation. In some video conferences, multiple speakers may discuss at the same time, so it is necessary to be able to intelligently identify the current speaker and ensure that only the audio stream of the speaker is processed. For example, in a corporate meeting with multiple people discussing, the audio sound pressure level (that is, the loudness of the speaker's voice) or other sound source localization technology may be used to identify who is the current speaker. By determining the audio stream of the main speaker, the server avoids processing other background sounds or irrelevant conversations, thereby improving the accuracy of subtitle generation.

[0061] The embodiment of the present invention can dynamically turn on or off the subtitle function according to user needs by obtaining a subtitle switch status message. This means that when subtitles are not needed, computing resources will not be wasted to process irrelevant audio streams, and subtitle generation will only be started when necessary. Not only does it improve the efficiency of the system, but it also effectively reduces the waste of computing resources and optimizes the performance of subtitle generation. By determining the audio stream according to the current speaker, it avoids processing irrelevant background noise or conversations of other participants. This ability to intelligently identify the speaker ensures that only voice information that really needs to be translated is processed, avoiding confusion or errors in the subtitle generation process. Therefore, the accuracy of the subtitle content is greatly improved, especially in a conference scene where multiple people are discussing, this method can effectively filter out interference sounds. Participants or conference hosts can control the opening or closing of subtitles through the client according to actual needs. This user-driven design improves the flexibility of the system, allowing each participant to customize the subtitle function according to their needs.

[0062] In an exemplary embodiment of the present invention, an implementation method of determining the audio stream based on a subtitle switch status message and a current speaker of a video conference is as follows: identifying identification information of a speaker in a subtitle-on state from a subtitle switch status message; determining whether there is a speaker corresponding to the speaker identification information among the current speakers of the video conference; and obtaining the audio stream if there is a speaker corresponding to the speaker identification information among the current speakers of the video conference.

[0063] During a video conference, when a participant decides to enable the subtitle function, the server will receive a corresponding "subtitle switch status message". This message is not just a simple switch signal, it also contains detailed speaker identification information, which is key data for identifying which speakers need to turn on subtitles. Speaker identification information usually includes the identity of the participant, such as name, user ID, role in the meeting, etc. For example, in a meeting, suppose there is a remote participant who needs real-time subtitles to follow the meeting progress. When the participant turns on the subtitle function through the client, the conference system will generate a subtitle switch status message, which clearly identifies the identity information of this participant. Through this identification information, the server can know that subtitles need to be generated for the speaker.

[0064] After receiving the subtitle switch status message, the server will check the current speaker in the video conference and determine whether there is a speaker that matches the speaker identification information in the subtitle switch status message. In other words, the server needs to verify whether the current speaker is the one who turned on the subtitle function before. This process is to avoid processing speakers who do not need subtitles and ensure that subtitles are generated only for speakers who need subtitles.

[0065] For example, during a discussion session of a meeting, it is detected that participants A and B are speaking, but the subtitle switch status message shows that only participant A has turned on the subtitle function. Therefore, only A's audio stream is judged whether it needs further processing, while the audio of participant B is ignored. This mechanism effectively reduces the burden of the system to process unnecessary information and improves resource utilization.

[0066] If it is determined that the current speaker is consistent with the speaker identification information in the subtitle switch status message, the server will further obtain the audio stream of the speaker. The audio stream refers to the real-time voice data of the speaker, which will be transmitted to the artificial intelligence subtitle terminal for voice recognition and subtitle translation. The core of this process lies in real-time and accuracy, that is, quickly extracting the speaker's voice from the conference audio stream and converting it into subtitles in the shortest time.

[0067] For example: suppose that in a meeting, a participant needs to use the subtitle function to understand the content of the speech. After receiving the subtitle switch status message, the server recognizes the identification information of the participant. Later, when the host of the meeting starts speaking, the server checks whether the current speaker matches the participant. If they match, it will immediately obtain the host's audio stream and send it to the artificial intelligence server for processing and subtitle generation. The generated subtitles are then transmitted to the participant's device to ensure that he can follow the meeting content in real time.

[0068] The embodiment of the present invention can accurately identify which speakers need subtitle generation by identifying the speaker identification information and judging the identity of the current speaker. This mechanism effectively avoids unnecessary audio processing, generates subtitles only for speakers who really need subtitles, reduces the burden on the system, and improves the accuracy of subtitle generation. Subtitle generation and processing require a certain amount of computing resources, especially when processing large video conferences. Through this implementation, the range of processed audio streams can be effectively controlled to avoid redundant audio processing for speakers who do not need to generate subtitles. According to the judgment of the subtitle switch status message, only the audio stream of a specific speaker is processed, thereby reducing excessive consumption of system computing resources. Allow participants to flexibly turn on or off the subtitle function according to their own needs. For example, some participants may be in a noisy environment and need subtitles to help understand the content of the meeting, while other participants may not need subtitles. Through this flexible control mechanism, the needs of different users can be met and the overall user experience of the video conference can be optimized. Since subtitles are only generated for clearly identified speakers, this implementation also improves the level of privacy protection to some extent. For example, if the content of some speakers is not suitable for public display or transcription into subtitles, these audio streams will not be processed, thereby protecting the privacy of the speakers.

[0069] In an exemplary embodiment of the present invention, a subtitle result is output to a participant terminal of a video conference so that the participant terminal displays the subtitle result. One implementation method is to transmit the subtitle result of a JavaScript Object Notation (JSON) structure to an artificial intelligence subtitle terminal; transmit the subtitle result to the participant terminal through an output channel of the artificial intelligence subtitle terminal, so that the participant terminal parses the subtitle result of the JSON structure to obtain the subtitle content, and displays the subtitle content according to the speaker corresponding to the subtitle result.

[0070] In actual application, the transmission of subtitle results uses the JSON format. JSON is a lightweight data exchange format that represents data in a simple key-value pair structure with good readability and cross-platform compatibility. Through the JSON structure, subtitle results can be transmitted in a unified format and are easy to be parsed by different systems. For example, a speaker spoke in a meeting, and the subtitles generated by the artificial intelligence server may contain the following information: the identity of the speaker, the time of the speech, the content of the subtitles, etc. This information will be encapsulated as a JSON object. By using the JSON structure, the server can ensure that different devices or platforms can easily read and parse the subtitle results. Regardless of whether the participants are using computers, mobile phones or other terminal devices, subtitle data in JSON format can be transmitted through standardized interfaces to ensure cross-device compatibility.

[0071] After generating the subtitle results, the server will transmit the subtitle results in JSON format to the participating terminals through the output channel of the artificial intelligence subtitle terminal. The artificial intelligence subtitle terminal is responsible for receiving and distributing the subtitle results. It establishes a stable data transmission channel with the participating terminals to ensure that the subtitles can be delivered in real time.

[0072] For example, in a meeting, when the conference host speaks, the AI ​​server will automatically generate subtitles and send them to the terminal devices of all participants through the output channel of the AI ​​subtitle terminal. Participants may be in different countries and use different devices, but because subtitles are transmitted in a unified JSON format, all devices can parse and display subtitles. This uniformity makes the delivery of subtitles more efficient and reliable.

[0073] In addition, the output channel can not only ensure the real-time transmission of subtitle results, but also optimize the transmission efficiency according to the network conditions, ensuring that subtitle data can be effectively transmitted even when the network is unstable. The artificial intelligence subtitle terminal has a certain cache capacity, which can save subtitle information when data transmission is interrupted and continue transmission after the network is restored.

[0074] After receiving the subtitle results in JSON structure, the participant terminal needs to parse it, obtain the subtitle content and display it on the user interface. The participant terminal generally contains a parsing module that specifically processes the JSON data transmitted from the artificial intelligence subtitle terminal. This module parses each field in the JSON, extracts the speaker and subtitle content, and displays the subtitle content to the user as needed.

[0075] Suppose that in a meeting, participants use different languages ​​and use the subtitle function provided by the system to understand the content of the speech. When the speaker turns on the subtitle function, the generated subtitles are automatically transmitted to all participants' devices via JSON. Each participant's terminal device will parse the subtitle data and display the corresponding speaker and subtitle content at the bottom of the screen to ensure that each participant can clearly see and understand the speaker's content.

[0076] The embodiment of the present invention uses the JSON structure to transmit subtitle results, ensuring that subtitle data can be easily parsed and displayed on different devices and operating systems. The simplicity and standardization of the JSON format make it an ideal choice for cross-platform communication, thereby improving the compatibility and applicability of the conference system. Through the speaker identification information in the JSON, the subtitles can be accurately associated with a specific speaker. Especially in multi-person conferences, subtitles can accurately show which speaker is speaking, avoiding the phenomenon of misalignment between subtitle content and speakers. Through the output channel of the artificial intelligence subtitle terminal, it can be ensured that the subtitle results are quickly transmitted to the participating terminals after generation. This real-time transmission mechanism enables participants to see the subtitle content instantly without any delays or lags.

[0077] In an exemplary embodiment of the present invention, after outputting the subtitle results to the participating terminals of the video conference so that the participating terminals display the subtitle results, an implementation method is to obtain a real-time subtitle closing message of the video conference; remove the artificial intelligence subtitle terminal from the video conference according to the real-time subtitle closing message, and notify the artificial intelligence server to release the artificial intelligence subtitle terminal.

[0078] In a video conference, participants can turn on or off the subtitle function at any time according to their needs. When a participant decides to turn off the subtitle function, the server will receive a subtitle off message. This message contains relevant information about the participant and the request to turn off the subtitle function. For example, in a meeting, a participant enables the subtitle function due to environmental noise or hearing impairment. As the meeting progresses, when the participant thinks that subtitles are no longer necessary, he or she can choose to turn off the subtitle function. At this time, the server will receive a real-time subtitle off message, indicating that the participant no longer needs subtitle service.

[0079] After receiving the real-time subtitles off message, the next step for the server is to remove the AI ​​subtitle terminal associated with the subtitle service from the video conference. This removal process can be understood as stopping the operation of the subtitle service and disconnecting the subtitle terminal from the video conference. For example, in a meeting with multiple speakers, some participants may have enabled the subtitle function early in the meeting to facilitate follow-up. However, as the meeting progressed, their needs changed and they decided to turn off the subtitles. In this case, the server automatically removes the AI ​​subtitle terminal from the meeting based on the subtitles off message. In this way, the conference system not only stops communicating with the terminal, but also avoids continuing to occupy computing resources when it is not necessary.

[0080] After removing the AI ​​subtitle terminal, the server will notify the AI ​​server to release the relevant subtitle terminal. The main purpose of this step is to ensure that system resources can be recycled in time to avoid redundant calculations and processing. The release of the subtitle terminal means that the computing resources and network connections occupied by the terminal will be recycled, and the AI ​​server can reallocate these resources to other participants or conference functions in need.

[0081] The embodiment of the present invention can effectively reduce unnecessary calculations and memory usage by automatically removing the AI ​​subtitle terminal and releasing resources after the real-time subtitle closing message. Real-time management of the subtitle function ensures that system resources are occupied only when needed, avoiding waste of resources, and can significantly reduce system load and improve system efficiency. The on and off state of the subtitle function usually changes dynamically, especially in long video conferences, where participants may frequently enable or disable the subtitle service as needed. By automatically removing subtitle terminals that are no longer in use and releasing resources, the risk of resource overload can be avoided, ensuring the normal operation of other functions (such as video and audio streams), and enhancing the overall stability of the system.

[0082] Based on the above description of the embodiment of a subtitle display method for a video conference, the following introduces a solution for displaying real-time subtitles of the speaker through AI voice recognition based on the visual network. Figure 2 , showing an interactive schematic diagram of a solution for displaying real-time subtitles of a speaker through AI voice recognition based on visual networking according to an embodiment of the present invention. The solution for displaying real-time subtitles of a speaker through AI voice recognition based on visual networking involves a conference management client (Pamir), a conference management server and an AI server.

[0083] Pamir supports configuring the main message number of the AI ​​server, and a new AI real-time subtitle function is added to the meeting: host / send one / send two / send three, four real-time subtitle switches.

[0084] The conference management server adds a new role type: AI real-time subtitle terminal (can decode 4 audio channels at the same time). When the conference management server receives the real-time subtitle activation message from Pamir, it transmits the conference ID to the AI ​​server through the main message number and applies for an AI subtitle terminal.

[0085] The conference management server adds the virtual terminal returned by the AI ​​server to the conference as the newly added AI real-time subtitle terminal. When receiving the message from the autonomous server that the AI ​​real-time subtitle terminal has successfully joined the conference, it determines which role's audio stream should be switched to the AI ​​real-time subtitle terminal based on the current speaker and the speaker subtitle switch status sent by Pamir. The decoding channel of the AI ​​real-time subtitle terminal defaults to the order of host / send one / send two / send three.

[0086] The AI ​​server transfers the received audio stream to the speech recognition interface for subtitle translation, and then converts the subtitle results of the four channels into a JSON structure and puts it into the output channel of the AI ​​real-time subtitle terminal.

[0087] The conference management server transfers the subtitle stream output by the AI ​​real-time subtitle terminal to all participating terminals. The participating terminals parse the JSON and display it on the screen according to different roles. When the speaker changes, the conference management server switches the audio stream of the new speaker to the channel corresponding to the AI ​​real-time subtitle terminal.

[0088] When the conference management server receives a message from Pamir to close the real-time subtitles for all roles, the conference management server removes the AI ​​real-time subtitle terminal that was previously added to the conference from the conference and notifies the AI ​​server to unbind the virtual terminal.

[0089] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0090] Reference Figure 3 , shows a structural block diagram of a subtitle display system for a video conference according to an embodiment of the present invention. The subtitle display system for a video conference is applied to a server, and the subtitle display system for a video conference may specifically include the following modules.

[0091] A subtitle-on message acquisition module 31 is used to acquire a real-time subtitle-on message of a video conference;

[0092] A subtitle terminal application module 32 is used to apply for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle opening message;

[0093] An audio stream transmission module 33, used to transmit the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle result of subtitle translation according to the speech recognition interface from the artificial intelligence server;

[0094] The subtitle result transmission module 34 is used to output the subtitle result to the participant terminal of the video conference so that the participant terminal can display the subtitle result.

[0095] In an exemplary embodiment of the present invention, the subtitle terminal application module 32 includes:

[0096] A conference identification reading module, used to read the conference identification information of the video conference from the real-time subtitle opening message;

[0097] A conference identification transmission module is used to transmit the conference identification information to the artificial intelligence server so that the artificial intelligence server allocates the artificial intelligence subtitle terminal to the video conference according to the conference identification information.

[0098] In an exemplary embodiment of the present invention, the system further comprises:

[0099] The subtitle terminal joining module is used to assign a subtitle terminal role to the artificial intelligence subtitle terminal and add the artificial intelligence subtitle terminal to the video conference before the audio stream transmission module 33 transmits the audio stream of the video conference to the artificial intelligence subtitle terminal.

[0100] In an exemplary embodiment of the present invention, the audio stream transmission module 33 includes:

[0101] A switch status acquisition module is used to obtain subtitle switch status messages from the client;

[0102] An audio stream determination module, configured to determine the audio stream according to the subtitle switch status message and the current speaker of the video conference;

[0103] An audio stream sending module is used to transmit the audio stream to the artificial intelligence subtitle terminal.

[0104] In an exemplary embodiment of the present invention, the audio stream determination module includes:

[0105] A speaker identification recognition module, used to identify the identification information of the speaker in the subtitle on state from the subtitle switch state message;

[0106] A speaker determination module, used for determining whether there is a speaker corresponding to the speaker identification information among the current speakers of the video conference;

[0107] The audio stream acquisition module is used to acquire the audio stream if there is a speaker corresponding to the speaker identification information among the current speakers of the video conference.

[0108] In an exemplary embodiment of the present invention, the subtitle result transmission module 34 includes:

[0109] A result sending module, used for transmitting the subtitle result in JSON structure to the artificial intelligence subtitle terminal;

[0110] The subtitle output module is used to transmit the subtitle result to the participant terminal through the output channel of the artificial intelligence subtitle terminal, so that the participant terminal parses the subtitle result of the JSON structure to obtain the subtitle content, and displays the subtitle content according to the speaker corresponding to the subtitle result.

[0111] In an exemplary embodiment of the present invention, the system further comprises:

[0112] A closing message acquisition module, configured to acquire a real-time closing message of the video conference after the subtitle result transmission module 34 outputs the subtitle result to the participating terminal of the video conference so that the participating terminal displays the subtitle result;

[0113] The subtitle terminal removal and release module is used to remove the artificial intelligence subtitle terminal from the video conference according to the real-time subtitle closing message, and notify the artificial intelligence server to release the artificial intelligence subtitle terminal.

[0114] As for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0115] 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.

[0116] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0117] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0118] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0120] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0121] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0122] The above is a detailed introduction to a subtitle display method for a video conference and a subtitle display system for a video conference 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 those 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. A subtitle display method for a video conference, characterized in that: Applied to a server, the method comprises: Get the real-time subtitle opening message of the video conference; Applying for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle opening message; Transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle result of subtitle translation according to the speech recognition interface from the artificial intelligence server; The subtitle result is output to a participant terminal of the video conference so that the participant terminal displays the subtitle result.

2. The method according to claim 1, characterized in that The applying for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle opening message includes: Reading the conference identification information of the video conference from the real-time subtitle opening message; The conference identification information is transmitted to the artificial intelligence server so that the artificial intelligence server allocates the artificial intelligence subtitle terminal to the video conference according to the conference identification information.

3. The method according to claim 1, characterized in that Before transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal, the method further includes: Assign a caption terminal role to the artificial intelligence caption terminal, and add the artificial intelligence caption terminal to the video conference.

4. The method according to claim 1, characterized in that: The step of transmitting the audio stream of the video conference to the artificial intelligence subtitle terminal includes: Get subtitle switch status message from the client; Determine the audio stream according to the subtitle switch status message and the current speaker of the video conference; The audio stream is transmitted to the artificial intelligence subtitle terminal.

5. The method according to claim 4, characterized in that The step of determining the audio stream according to the subtitle switch status message and the current speaker of the video conference includes: Identifying identification information of a speaker in a subtitle-on state from the subtitle switch state message; Determine whether there is a speaker corresponding to the speaker identification information among the current speakers of the video conference; If there is a speaker corresponding to the speaker identification information among the current speakers of the video conference, the audio stream is obtained.

6. The method according to claim 1, characterized in that Outputting the subtitle result to a participant terminal of the video conference so that the participant terminal displays the subtitle result includes: Transmit the subtitle result of the JSON structure to the artificial intelligence subtitle terminal; The subtitle result is transmitted to the participant terminal through the output channel of the artificial intelligence subtitle terminal, so that the participant terminal parses the subtitle result of the JSON structure to obtain the subtitle content, and displays the subtitle content according to the speaker corresponding to the subtitle result.

7. The method according to claim 1, characterized in that After outputting the subtitle result to a participant terminal of the video conference so that the participant terminal displays the subtitle result, the method further includes: Obtaining a real-time subtitle closing message for the video conference; The artificial intelligence subtitle terminal is removed from the video conference according to the real-time subtitle closing message, and the artificial intelligence server is notified to release the artificial intelligence subtitle terminal.

8. A subtitle display system for video conferencing, characterized in that: Applied to a server, the system comprises: A subtitle-on message acquisition module is used to acquire real-time subtitle-on messages of a video conference; A subtitle terminal application module, used to apply for an artificial intelligence subtitle terminal from an artificial intelligence server according to the real-time subtitle opening message; An audio stream transmission module, used to transmit the audio stream of the video conference to the artificial intelligence subtitle terminal, so that the artificial intelligence subtitle terminal transmits the audio stream to the artificial intelligence server, and receives the subtitle result of subtitle translation according to the speech recognition interface from the artificial intelligence server; The subtitle result transmission module is used to output the subtitle result to the participant terminal of the video conference so that the participant terminal can display the subtitle result.

9. An electronic device, characterized in that: include: one or more processors; and One or more machine-readable media having instructions stored thereon, when executed by the one or more processors, enable the electronic device to execute the subtitle display method for video conferencing as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer program stored therein enables the processor to execute the subtitle display method for a video conference as described in any one of claims 1 to 7.

Images