Method and device for improving utilization rate of digital human

By caching videos in advance and calling them on demand, the problem of low digital human utilization and high-resolution video transcoding is solved, and more efficient digital human utilization and media server load is achieved.

CN119071557APending Publication Date: 2024-12-03SHENZHEN THINKIVE INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411300247.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-18
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The utilization rate of digital people is too low, mainly because commercial digital people charge according to the number of connected lines, which limits the concurrency of account opening videos, and high-resolution video transcoding puts pressure on media servers.

Method used

By cache videos containing variables in advance during the account opening process, if the broadcasting technique contains variables, switch to pre-cache videos, otherwise switch to digital human real-time synthesizing videos, reduce real-time rendering and generation, and realize on-demand calls.

Benefits of technology

It greatly reduces the call frequency of real-time synthesis of digital people, improves the utilization rate of digital people, and reduces the pressure on media servers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119071557A_ABST
    Figure CN119071557A_ABST
Patent Text Reader

Abstract

The invention discloses a method and device for improving the utilization rate of a digital human, and the method comprises the steps: firstly playing a silent video in response to a received digital human broadcast request, and obtaining a broadcast verbal skill according to the digital human video broadcast request; and checking whether a variable is included in the broadcast verbal skill, if the variable is included, checking whether a video related to the variable is cached in a pre-cached video, if so, switching the silent video to the pre-cached video, and if not, switching the silent video to a video synthesized by a digital human in real time. According to the method provided by the invention, the video containing the variable is cached in advance in the account opening process without real-time rendering generation, and the pre-cached video is utilized to respond to the interaction application of the user, so that the calling of the real-time synthesis of the digital human in the service is greatly reduced, the calling on demand is realized, and the user experience is improved. Therefore, the utilization rate of the digital human is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computers, and in particular to a method and device for improving the utilization rate of digital humans. Background Art

[0002] Digital human refers to a virtual human image that is cloned from a real person's image and voice through artificial intelligence technology. Compared with the traditional account opening model, digital human account opening assists in completing the entire process of account opening video witnessing, and can use interactive forms such as simulated image and voice to replace simple text prompts, so as to better answer and guide customers' account opening operations, providing a more intimate, natural and efficient account opening experience.

[0003] However, the utilization rate of digital humans in the conventional way is too low, which is mainly reflected in the following aspects:

[0004] 1. Commercial digital humans are usually charged based on the number of connection lines, and one account opening video occupies one digital human connection line throughout the entire process. Based on cost considerations, users will strictly control the number of digital human connection lines, thereby limiting the number of concurrent account opening videos.

[0005] 2. When the digital human video is pushed to the front end, it will be transcoded once (decode the digital human video into the original format, and then re-encode it according to the encoding format required for video transmission). When the video resolution is high, the pressure on the media server on the transcoding side will be very high. Summary of the invention

[0006] To this end, the present invention aims to solve the deficiencies in the prior art to at least a certain extent, thereby proposing a method and device for improving the utilization rate of digital humans.

[0007] In a first aspect, the present invention provides a method for improving the utilization rate of digital humans, the method comprising:

[0008] In response to the received digital human broadcast request, firstly play the silent video and obtain the broadcasting words according to the digital human video playback request;

[0009] Check whether the broadcast script contains variables. If so, check whether there is a video related to the variable cached in the pre-cached video. If so, switch the silent video to the pre-cached video. If not, switch the silent video to a video synthesized in real time by a digital human.

[0010] In a second aspect, the present invention provides a device for improving the utilization rate of a digital human, the device comprising:

[0011] Playback module: used to respond to the received digital human broadcast request, first play the silent video and obtain the broadcasting words according to the digital human video playback request;

[0012] Switching module: used to check whether the broadcast script contains variables. If so, check whether there is a video related to the variable cached in the pre-cached video. If so, switch the silent video to the pre-cached video. If not, switch the silent video to a video synthesized in real time by a digital human.

[0013] In a third aspect, the present invention further provides a device for improving the utilization rate of digital humans, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps in the method for improving the utilization rate of digital humans as described in the first aspect.

[0014] In a fourth aspect, the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for improving the utilization rate of digital humans as described in the first aspect are implemented.

[0015] The present invention provides a method and device for improving the utilization rate of digital human, the method comprising: in response to a received digital human broadcast request, first playing a silent video and obtaining a broadcast script according to the digital human video playback request; checking whether the broadcast script contains a variable, if it contains a variable, checking whether a video related to the variable is cached in a pre-cached video, if yes, switching the silent video to the pre-cached video, if not, switching the silent video to a video synthesized by a digital human in real time. Through the method provided by the present invention, by caching the video containing the variable in advance during the account opening process, there is no need for real-time rendering and generation, and using the pre-cached video to respond to the user's interactive application by introduction, the call for real-time synthesis of the digital human in the service will be greatly reduced, and on-demand calling will be realized, thereby improving the utilization rate of the digital human. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0017] Figure 1 A schematic diagram of a flow chart of a method for improving the utilization rate of digital humans according to the present invention;

[0018] Figure 2 A schematic diagram of a sub-flow of the method for improving the utilization rate of digital human according to the present invention;

[0019] Figure 3It is another sub-flow diagram of the method for improving the utilization rate of digital human according to the present invention;

[0020] Figure 4 It is another sub-flow diagram of the method for improving the utilization rate of digital human of the present invention;

[0021] Figure 5 A schematic diagram of the digital human video interaction timing of the method for improving the utilization rate of digital humans of the present invention;

[0022] Figure 6 It is another sub-flow diagram of the method for improving the utilization rate of digital human of the present invention;

[0023] Figure 7 A schematic diagram of the method for improving the utilization rate of digital humans before video transcoding is improved;

[0024] Figure 8 It is a schematic diagram of the method for utilizing the avatar digital human of the present invention before the improvement of video transcoding;

[0025] Fig. 9 A test image of conventional video transcoding of the method for improving the utilization rate of digital humans of the present invention;

[0026] Fig.10 A test chart of code stream format transcoding of the method for improving the utilization rate of digital human of the present invention;

[0027] Fig.11 It is another sub-flow diagram of the method for improving the utilization rate of digital human of the present invention;

[0028] Fig.12 It is a schematic diagram of the overall structure of the method for improving the utilization rate of digital human according to the present invention;

[0029] Fig.13 It is a schematic diagram of program modules of the device for improving the utilization rate of digital human according to the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, features and advantages of the invention more obvious and easy to understand, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0031] Please refer to Figure 1 , Figure 1 A flow chart of a method for improving the utilization rate of a digital human is provided for the present invention in the embodiment of the present application. In this embodiment, the method for improving the utilization rate of a digital human includes:

[0032] Step 101: In response to a received digital human broadcast request, first play a silent video and obtain the broadcast script according to the digital human video playback request.

[0033] In this embodiment, during the account opening process, after receiving the digital human broadcast request and responding, the silent video is played first, and the broadcast script is obtained according to the digital human broadcast request. The digital human video status includes silent state and broadcast state. In the silent state, the digital human video is muted and only repeated expressions and actions are played in a loop. In the broadcast state, the digital human video plays sound, and the digital human's body language and voice will change with the changes in the broadcast content.

[0034] Step 102, checking whether the broadcast script contains variables, if so, checking whether there is a video related to the variable cached in the pre-cached video, if yes, switching the silent video to the pre-cached video, if not, switching the silent video to a video synthesized in real time by a digital human.

[0035] In this embodiment, in the account opening scenario, the broadcast scripts are relatively fixed. Except for a few variable scripts with customer names, most of the general scripts are fixed.

[0036] First, the broadcast script is obtained according to the digital human playback request, and then the broadcast script is checked to see if it contains a variable. If the broadcast script contains a variable, the variable is the customer's name. When the silent video needs to be switched to a real-time synthesized video, first check whether there is a video related to the customer's name in the pre-cached video. If there is a related video in the pre-cached video, the silent video is switched to the pre-cached video related to the variable, wherein the pre-cached video includes multiple videos that have been cached in advance. If there is no video related to the variable in the pre-cached video, the silent video is switched to a video synthesized in real time by the digital human, that is, the silent video is switched to a real-time synthesized video and played.

[0037] Because commercial digital humans are usually charged according to the number of connection lines, and an account opening video occupies one digital human connection throughout the process, users will strictly control the number of digital human connection lines, thereby limiting the concurrent number of account opening videos. By caching multiple variable-related videos in advance, the calls for real-time synthesis of digital humans in the service can be greatly reduced, and on-demand calls can be achieved through caching, thereby improving the utilization rate of digital humans.

[0038] The embodiment of the present application provides a method for improving the utilization rate of digital human, the method comprising: in response to the received digital human broadcast request, first playing a silent video and obtaining the broadcast script according to the digital human video playback request; checking whether the broadcast script contains variables, if it contains variables, checking whether there is a video related to the variable cached in the pre-cached video, if yes, switching the silent video to the pre-cached video, if not, switching the silent video to a video synthesized by the digital human in real time. Through the method provided by the present invention, by caching the video containing the variable in advance during the account opening process, there is no need for real-time rendering and generation, and using the pre-cached video to respond to the user's interactive application using the introduction, the call for real-time synthesis of the digital human in the service will be greatly reduced, and on-demand calling will be realized, thereby improving the utilization rate of the digital human.

[0039] For further information, see Figure 2 , Figure 2 The sub-flow diagram of the method for improving the utilization rate of digital human in the embodiment of the present application is as follows. In the embodiment, if the variable is included, it is checked whether the pre-cached video has a video related to the variable cached therein, if so, the silent video is switched to the pre-cached video, if not, the silent video is switched to a video synthesized in real time by the digital human, including:

[0040] Step 201: If the broadcast script includes the customer's name, first search in the pre-cached Hundred Family Names video, wherein the variable at least includes the customer's name, and the pre-cached video includes the pre-cached Hundred Family Names video;

[0041] Step 202: If no video is found, the digital human interface is called to drive the digital human to synthesize the video in real time and play it.

[0042] In this embodiment, if the acquired broadcast script contains variables, where the variables are generally the customer's name, that is, when the broadcast script contains the customer's name, it is necessary to switch the silent video to the real-time synthesized video. When it is necessary to switch to the real-time synthesized video, first query whether there is a cached video related to the customer's name in the pre-cached video, where the pre-cached video includes the pre-cached Hundred Family Names video, and save the pre-cached Hundred Family Names video to the local cache file, such as Miss Zhang, Mr. Wang, Ms. Wu, etc. That is, query whether there is a cached video related to the customer's name in the pre-cached Hundred Family Names video. If there is, the cached video related to the customer's name is played. If not, the digital human interface is called to drive the digital human to synthesize the video related to the customer's name in real time, and the real-time synthesized video is played.

[0043] For further information, see Figure 3 , Figure 3The sub-flow diagram of the method for improving the utilization rate of digital human in the embodiment of the present application is shown in FIG. In the embodiment, if the digital human is not available, the silent video is switched to a video synthesized in real time by the digital human, and the method further includes:

[0044] Step 301: Save the video synthesized in real time by the digital human into the pre-cached video and create an index;

[0045] Step 302: If the broadcasting script contains the variable, searching the pre-cached videos for videos related to the variable through the index, and playing the pre-cached videos;

[0046] Step 303: If no video is found, the real-time synthesized video of the digital human is called to play, and the real-time synthesized video is saved in the pre-cached video.

[0047] In this embodiment, each time a video that needs to be synthesized in real time by driving the digital human through calling the digital human interface is saved, it is saved in the pre-cached video, that is, the pre-cached video not only stores the pre-cached Hundred Family Names videos, but also stores the video synthesized in real time by the digital human. That is, when there is no video related to the customer's name in the pre-cached video, the digital human is used to perform real-time synthesis, and the real-time synthesized video is saved, and the video synthesized and saved by the digital human in real time is indexed.

[0048] When the broadcast script contains the customer's name, the video synthesized in real time by the digital human can be indexed to determine whether there is a video related to the customer's name. If so, the video related to the customer's name saved in the pre-cached video will be played.

[0049] If there is no video related to the customer's name in the pre-cached videos, the digital human interface is called to drive the digital human to play the real-time synthesized video, and the real-time synthesized video is saved in the pre-cached videos so that the real-time synthesized video can be used again next time without the need for real-time synthesis, thereby reducing the call for real-time synthesis of digital humans during the account opening process.

[0050] For further information, see Figure 4 and Figure 5 , Figure 4 Schematic diagram of a sub-flow of a method for improving the utilization rate of a digital human in an embodiment of the present application. Figure 5 The following is a schematic diagram of the timing of the digital human video interaction in an embodiment of the present application. In this embodiment, the response to the received digital human broadcast request also includes:

[0051] Step 401: Transmitting the digital human broadcast request to the video server through the H5 front end;

[0052] Step 402: Based on the video server, the AI ​​video server is connected, and an interface is called to drive the digital human and generate a digital human video stream in real time;

[0053] Step 403: Push the digital human video stream generated in real time to the video server through the AI ​​video server, transcode the digital human video stream through the video server and push it to the H5 front end.

[0054] In this embodiment, the H5 front end first transmits the digital human broadcast request to the video server. After the video server receives the digital human broadcast request from the H5 front end, it transfers the digital human broadcast request to the AI ​​video server, and drives the digital human to generate a digital human video stream by calling the relevant interface. The AI ​​video server pushes the real-time generated digital human video stream to the video server. After the video server receives the digital human video stream, it transcodes the received digital human video stream and pushes the transcoded digital human video stream to the H5 front end. The video server and the AI ​​server push steps are repeated until the account opening process is completed. Among them, the AI ​​server includes a voice server and a video server.

[0055] For further information, see Figure 6 , Figure 6 The following is a sub-flow diagram of a method for improving the utilization rate of digital human in an embodiment of the present application. In this embodiment, the video server includes a control server and a media server. The bitstream format conversion of the digital human video stream by the video server includes:

[0056] Step 501: The format of the digital human video stream generated by the AI ​​video server is an H264 code stream in avcc mode;

[0057] Step 502: Based on the media server, the digital human video stream is converted from an H264 code stream in avcc mode to an H264 code stream in annexb mode, and pushed to the H5 front end.

[0058] In this embodiment, the video server includes a control server and a media server. When the H5 front end sends a digital human broadcast request to the control server, the control server sends the digital human broadcast request to the media server. The media server will connect to the AI ​​video server and call the relevant interface to drive the digital human to generate a digital human video stream. The AI ​​video server pushes the real-time generated digital human video stream to the media server, and the media server transcodes the received digital human video stream and pushes it to the H5 front end.

[0059] See also Figure 7 and Figure 8 , Figure 7This is a schematic diagram of the video transcoding before improvement of the method for improving the utilization rate of digital human in the embodiment of the present application, Figure 8 This is a schematic diagram of the improved video transcoding of the method for improving the utilization rate of digital human in the embodiment of the present application. Before the improvement, the H264 code stream in avcc mode needs to be decoded into pixel data in YUV format before it can be converted into the H264 code stream in annexb mode. After the improvement, the digital human video stream is directly converted from the H264 code stream in avcc mode to the H264 code stream in annexb mode. Figure 7 and Figure 8 The difference is that in the improved conversion process, the video is no longer transcoded (decoded first and then encoded), but instead the bitstream format is converted, converting the H264 bitstream from avcc mode to annexb mode.

[0060] When the media server pushes the digital human video stream to the H5 front end, it will first perform a transcoding operation, that is, first decode the digital human video into the original format, and then re-encode it according to the encoding format required for video transmission. When the video resolution is high, the pressure on the media server on the transcoding side will be very high. Therefore, through the bitstream format conversion, the digital human video stream no longer needs to be transcoded. Compared with conventional video transcoding, bitstream format conversion will occupy much less resources; and audio transcoding itself occupies fewer resources.

[0061] The video is converted through the bitstream format, so there is no need to decode the video, and there is no need to allocate memory to store the decoded video data. In addition, the video data in the original YUV format is much larger than the video data in the H264 bitstream format. Therefore, the memory usage will be significantly reduced by using the bitstream format conversion mode.

[0062] After the above adjustments, the pressure on the media server on the transcoding side is effectively reduced. When playing videos on the media server, please refer to Fig. 9 and Fig.10 , Fig. 9 This is a test image of conventional video transcoding of the method for improving the utilization rate of digital humans in the embodiment of the present application. Fig.10 This is a test image of code stream format transcoding of the method for improving the utilization rate of digital human in the embodiment of the present application. Fig. 9 and Fig.10 By comparison, it can be seen that the CPU consumption of the bitstream format conversion solution is only 1 / 9 of that of the conventional video transcoding solution, and the memory usage of the bitstream format conversion solution is only 1 / 4 of that of the conventional transcoding solution.

[0063] Specifically, the audio and video encoding formats supported by WebRTC can be: OPUS, H264; and the audio and video encoding formats of MP4 files and digital human videos are usually: AAC, H264, so the H264 stream can be converted from avcc mode to annexb mode through stream format conversion, where avcc mode is used for video files and annexb mode is used for network transmission. In annexb mode, the data (NALUnit) of each video frame uses 0x000001 or 0x0000001 as a delimiter, and key information such as SPS (Sequence Parameter Set) and PPS (Picture Parameter Set) usually appears periodically in the video stream, especially before the key frame (I frame).

[0064] For further information, see Fig.11 , Fig.11 The following is a sub-flow diagram of a method for improving the utilization rate of digital humans in an embodiment of the present application. In this embodiment, the digital human broadcast request is transmitted to the video server through the H5 front end, including:

[0065] Step 601: Perform protocol interaction with the control server through the H5 front end, and transmit the digital human broadcast request to the control server;

[0066] Step 602: Transmit the digital human broadcast request to the media server based on the control server.

[0067] In this embodiment, the video server includes a control server and a media server. The H5 front end is connected to the control server to perform protocol interaction. The content of the protocol interaction includes: logging into the system, entering the room, etc., and establishing a media channel with the media server. While the H5 front end performs protocol interaction with the control server, it also transmits the digital human broadcast request to the control server, and the control server then transmits the received digital human broadcast request to the media server.

[0068] Furthermore, the checking whether the broadcasting script contains variables further includes:

[0069] If the broadcast script does not contain variables, the silent video will be switched to a fixed script video until the fixed script video is finished playing, and then switched to the silent video.

[0070] In this embodiment, in step 102, when checking whether the broadcast script contains variables, if the broadcast script does not contain variables, the silent video is switched to the fixed script video until the fixed script video is played, and then the video is switched to the silent video, wherein the fixed script video can reduce the waste of computing resources for real-time synthesis of digital humans. Among them, digital human videos are divided into three categories: silent video, fixed script video, and real-time synthesized video. By default, silent video and fixed script video are both locally pre-recorded MP4 files, and real-time synthesized video needs to call the digital human interface to obtain it in real time.

[0071] See also Fig.12 , Fig.12 This is a schematic diagram of the overall structure of the method for improving the utilization rate of digital human by introducing cached video in an embodiment of the present application. The specific steps are as follows:

[0072] 1. During the account opening process, when the H5 front-end receives the video, the default silent video is played first.

[0073] 2. When the requested script does not contain variables (customer name), the default silent video will be switched to a fixed script video.

[0074] 3. After the fixed-speech video is played, switch back to the default silent video.

[0075] 4. When the requested broadcast script contains variables (customer name), the default silent video will be switched to real-time composite video.

[0076] 5. After the real-time composite video is played, switch back to the default silent video.

[0077] 6. Repeat steps 2-5 until the account opening process is completed.

[0078] Furthermore, the specific steps of the method for improving the utilization rate of digital humans by reducing the pressure on the media server at the transcoding side are as follows:

[0079] 1. The H5 front end connects to the video control server to perform protocol interaction, including logging into the system, entering the room, etc., and establishing a media channel with the media server.

[0080] 2. After receiving the broadcast request from the H5 front-end digital human, the control server forwards the request to the media server.

[0081] 3. The media server connects to the AI ​​video server and calls relevant interfaces to drive the digital human.

[0082] 4. The AI ​​video server pushes the real-time generated digital human video stream to the media server.

[0083] 5. The media server transcodes the received digital human video stream through the bitstream format, that is, converts the digital human video stream in the H264 bitstream format from avcc mode to annexb mode, and pushes it to the H5 front end.

[0084] Furthermore, the embodiment of the present application also provides a device 700 for improving the utilization rate of a digital human. Fig.13 Schematic diagram of program modules of the device for improving the utilization rate of digital human in the embodiment of the present application. In the present embodiment, the device 700 for improving the utilization rate of digital human includes:

[0085] Playing module 701: used for responding to the received digital human broadcast request, first playing the silent video and obtaining the broadcasting words according to the digital human video playback request;

[0086] Switching module 702: used to check whether the broadcast script contains variables. If so, check whether there is a video related to the variable cached in the pre-cached video. If so, switch the silent video to the pre-cached video. If not, switch the silent video to a video synthesized in real time by a digital human.

[0087] The embodiment of the present application provides a device for improving the utilization rate of digital human, which can achieve: in response to the received digital human broadcast request, first play a silent video and obtain the broadcast script according to the digital human video playback request; check whether the broadcast script contains variables, if it contains variables, check whether there is a video related to the variable cached in the pre-cached video, if yes, switch the silent video to the pre-cached video, if not, switch the silent video to the video synthesized by the digital human in real time. Through the method provided by the present invention, by caching the video containing variables in advance during the account opening process, there is no need for real-time rendering and generation, and using the pre-cached video to respond to the user's interactive application using the introduction, it will greatly reduce the call for real-time synthesis of digital humans in the service, thereby improving the utilization rate of digital humans.

[0088] Furthermore, the present application also provides a device for improving the utilization rate of digital humans, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the various steps in the method for improving the utilization rate of digital humans as described above are implemented.

[0089] Furthermore, the present application also provides a storage medium on which a computer program is stored. When the computer program is executed, each step in the method for improving the utilization rate of digital humans as described above is implemented.

[0090] Each functional module in each embodiment of the present invention may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may be stored in a computer-readable storage medium.

[0091] Based on this understanding, the technical solution of the specification of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., various media that can store program codes.

[0092] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present invention is not limited by the described order of actions, because according to 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 and modules involved are not necessarily required by the present invention. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0093] For those skilled in the art, according to the concept of the embodiments of the present application, there may be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for improving the utilization rate of digital human, characterized in that: The method comprises: In response to the received digital human broadcast request, firstly play the silent video and obtain the broadcasting words according to the digital human video playback request; Check whether the broadcast script contains variables. If so, check whether there is a video related to the variable cached in the pre-cached video. If so, switch the silent video to the pre-cached video. If not, switch the silent video to a video synthesized in real time by a digital human.

2. The method according to claim 1, characterized in that If the variable is included, checking whether a video related to the variable is cached in the pre-cached video, if yes, switching the silent video to the pre-cached video, if no, switching the silent video to a video synthesized in real time by a digital human, including: If the broadcast script contains the customer's name, a query is first performed in the pre-cached Hundred Family Names video, wherein the variable at least includes the customer's name, and the pre-cached video includes the pre-cached Hundred Family Names video; If no result is found, the digital human interface is called to drive the digital human to synthesize the video in real time and play it.

3. The method according to claim 2, characterized in that If not, the silent video is switched to a video synthesized in real time by a digital human, further comprising: Saving the video synthesized in real time by the digital human into the pre-cached video and creating an index; If the broadcasting script contains the variable, searching the pre-cached videos for videos related to the variable through the index, and playing the pre-cached videos; If no result is found, the real-time synthesized video of the digital human is called to play, and the real-time synthesized video is saved in the pre-cached video.

4. The method according to claim 1, characterized in that: The step of responding to the received digital human broadcast request further includes: The digital human broadcast request is transmitted to the video server through the H5 front end; Based on the video server connected to the AI ​​video server, the interface is called to drive the digital human and generate a digital human video stream in real time; The digital human video stream generated in real time is pushed to the video server through the AI ​​video server, and the digital human video stream is transcoded and pushed to the H5 front end through the video server.

5. The method according to claim 4, characterized in that The video server includes a control server and a media server, and the bitstream format conversion of the digital human video stream by the video server includes: The format of the digital human video stream generated by the AI ​​video server is an H264 code stream in avcc mode; Based on the media server, the digital human video stream is converted from the H264 code stream in the avcc mode to the H264 code stream in the annexb mode, and pushed to the H5 front end.

6. The method according to claim 3, characterized in that The step of transmitting the digital human broadcast request to the video server through the H5 front end includes: Perform protocol interaction with the control server through the H5 front end, and transmit the digital human broadcast request to the control server; The digital human broadcast request is transmitted to a media server based on the control server.

7. The method according to claim 1, characterized in that The checking whether the broadcasting script contains variables also includes: If the broadcast script does not contain variables, the silent video will be switched to a fixed script video until the fixed script video is finished playing, and then switched to the silent video.

8. A device for improving the utilization rate of digital human, characterized in that: The device comprises: Playback module: used to respond to the received digital human broadcast request, first play the silent video and obtain the broadcasting words according to the digital human video playback request; Switching module: used to check whether the broadcast script contains variables. If so, check whether there is a video related to the variable cached in the pre-cached video. If so, switch the silent video to the pre-cached video. If not, switch the silent video to a video synthesized in real time by a digital human.

9. A device for improving the utilization rate of digital humans, 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, each step of the method for improving the utilization rate of digital humans as described in any one of claims 1 to 7 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the method for improving the utilization rate of digital humans as described in any one of claims 1 to 7 is implemented.