阅读说明：本技术 基于在线教育的视频生成方法、系统、设备及存储介质 (Video generation method, system, device and storage medium based on online education ) 是由 孙高朝 涂序文 于 2020-03-20 设计创作，主要内容包括：本发明提供了基于在线教育的视频生成方法、系统、设备及存储介质,其中,方法包括在授课的过程中,采集第一类用户的具有时间戳的第一类视频与每个第二类用户的具有时间戳的第二类视频；自第一类视频中删除不发言的视频段落获得第一类子视频,自每个第二类视频中删除不发言的视频段落获得第二类子视频；自第一类子视频中筛选符合预设主题语码的第一类子视频作为第一目标视频；自所有第二类用户的时间戳位于第一目标视频之前且最相邻的一个第二类子视频作为第二目标视频；剪辑获得对应提供第二类子视频的第二类用户的主题对话视频,本发明能够根据老师与学员的对话,自动剪辑反应学员关于主题对话视频,提高人性化体验。(The invention provides a video generation method, a system, equipment and a storage medium based on online education, wherein the method comprises the steps of collecting a first type video with a timestamp of a first type user and a second type video with a timestamp of each second type user in the course of teaching; deleting the video paragraphs which do not speak from the first type of video to obtain first type of sub-videos, and deleting the video paragraphs which do not speak from each second type of video to obtain second type of sub-videos; screening a first-class sub-video which accords with a preset theme language code from the first-class sub-video to serve as a first target video; taking a second-class sub-video which is positioned before the first target video and is nearest to the first target video from the timestamps of all the second-class users as a second target video; the method and the system can automatically clip the theme dialogue video of the student about the theme dialogue according to the dialogue between the teacher and the student, and improve the humanized experience.)

1. A video generation method based on online education is characterized by comprising the following steps:

s110, collecting a first type video with a time stamp of a first type user and a second type video with a time stamp of each second type user in the process that at least one first type user gives lessons to a plurality of second type users;

s120, deleting video paragraphs which are not spoken by the first class of users from the first class of videos to obtain a plurality of sections of first class sub-videos which are arranged according to a time sequence, deleting video paragraphs which are not spoken by the second class of users from each second class of video to obtain a plurality of sections of second class sub-videos which are arranged according to the time sequence;

s130, screening the first-class sub-videos which accord with at least one preset theme language code in a preset theme language code set from the first-class sub-videos to serve as first target videos;

s140, taking a second type sub-video which is positioned before the first target video and is nearest to the first target video from the time stamps of all the second type users as a second target video;

s150, obtaining the theme dialogue video corresponding to the second type of users providing the second type of sub-videos based on at least the time sequence cutting of the second target video and the first target video.

2. The method for generating video based on online education as claimed in claim 1, wherein in step S110, each frame of the first type of video has a time stamp generated in real time when shooting;

and each frame of the second-class video of each second-class user is provided with a time stamp generated in real time during shooting.

3. The method for generating a video based on online education as claimed in claim 1, wherein in step S120, a video segment with an audio decibel below a preset threshold is deleted from the first type of video as a video segment for which the first type of user does not speak;

and taking the video section with the audio decibel number lower than a preset threshold value in each second type of video as a video section for which the second type of user does not speak, and deleting the video section from the second type of video.

4. The method for generating videos based on online education as claimed in claim 1, wherein in step S130, the set of preset subject language codes includes a plurality of voices expressing the evaluated preset subject language codes, and the voices satisfying at least one preset subject language code are searched for as the first target video in the first class of sub-videos through voice similarity comparison.

5. The method for generating video based on online education as claimed in claim 1, wherein in step S130, the set of preset topic language codes includes a plurality of words of the preset topic language codes for expressing the evaluation, and after the first type of sub-video is converted into text by voice recognition, the words satisfying at least one of the preset topic language codes are searched as the first target video.

6. The method for generating video based on online education as claimed in claim 1, wherein in step S130, the preset topic language code set has a plurality of preset topic language codes expressing praise.

7. The method for generating a video based on online education as claimed in claim 1, wherein in step S130, the preset subject code set has a plurality of preset subject code expressing criticism.

8. The method for generating video based on online education as claimed in claim 1, wherein in step S140, one of the second type sub-videos of all the second type users whose timestamp of the ending frame satisfying the second type sub-video is before the timestamp of the starting frame of the first target video and whose time interval from the starting frame is the smallest is searched for as the second target video.

9. The video generating method based on online education as recited in claim 1, wherein in step S150, one of the first-class sub-videos that is located before and closest to the first target video is obtained as a third target video, and the third target video, the second target video, and the first target video are clipped together in time sequence as a topic conversation video of the second-class user corresponding to the second-class sub-video.

10. A method for video generation based on online education as claimed in claim 1 wherein at least one topic conversation video of the second class of users is sent to at least one third class of users associated with the second class of users.

11. The video generating method based on online education as claimed in claim 1, wherein at least one topic conversation video of the second class of users and the course information are generated into a web of course advertisement information to be transmitted to at least one third class of users associated with the second class of users.

12. A video generation system based on online education for implementing the video generation method based on online education claimed in claim 1, characterized by comprising:

the video acquisition module is used for acquiring a first type video with a timestamp of a first type user and a second type video with a timestamp of each second type user in the process of giving lessons to a plurality of second type users by at least one first type user;

the video segmentation module deletes video paragraphs which are not spoken by the first class users from the first class video to obtain a plurality of segments of first class sub-videos which are arranged according to a time sequence, deletes video paragraphs which are not spoken by the second class users from each second class video to obtain a plurality of segments of second class sub-videos which are arranged according to the time sequence;

the video screening module is used for screening the first-class sub-videos which accord with at least one preset theme language code in a preset theme language code set from the first-class sub-videos to serve as first target videos;

the video positioning module is used for taking a second type of sub-video which is positioned before the timestamp of the first target video and is nearest to the first target video from the timestamps of all the second type of users as a second target video;

and the video generation module is used for obtaining the theme conversation video corresponding to the second type of users providing the second type of sub-videos at least based on the time sequence cutting of the second target video and the first target video.

13. The video generation system based on online education of claim 12, wherein each frame of the first type of video has a time stamp generated in real time when taken;

and each frame of the second-class video of each second-class user is provided with a time stamp generated in real time during shooting.

14. The online education-based video generation system of claim 12, wherein the video segmentation module deletes, from the first type of video, a video segment whose audio decibel number is lower than a preset threshold as a video segment for which the first type of user does not speak;

and taking the video section with the audio decibel number lower than a preset threshold value in each second type of video as a video section for which the second type of user does not speak, and deleting the video section from the second type of video.

15. The system of claim 12, wherein the set of predetermined subject codes comprises a plurality of voices expressing evaluated predetermined subject codes, and the first type of sub-video searches for a voice satisfying at least one predetermined subject code as the first target video by comparing voice similarities.

16. The video generation system based on online education as claimed in claim 12, wherein the set of predetermined subject language codes includes a plurality of words expressing the predetermined subject language codes for evaluation, and after the first type of sub-video is converted into a text by speech recognition, words satisfying at least one predetermined subject language code are searched as the first target video.

17. The video generation system based on online education of claim 12, wherein in the video positioning module, one of the second type sub-videos of all the second type users whose timestamp of the ending frame satisfying the second type sub-video is before the timestamp of the starting frame of the first target video and whose time interval from the starting frame is the smallest is searched as the second target video.

18. The video generation system based on online education of claim 12, wherein in the video generation module, one of the first-class sub-videos that is located before and closest to the first target video is obtained as a third target video, and the third target video, the second target video, and the first target video are chronologically clipped together as a topic conversation video of the second-class user to which the second-class sub-video corresponds.

19. A video generating device based on online education, comprising:

a processor;

a memory having stored therein executable instructions of the processor;

wherein the processor is configured to perform the steps of the online education-based video generation method of any one of claims 1 to 11 via execution of the executable instructions.

20. A computer-readable storage medium storing a program, wherein the program when executed implements the steps of the online education-based video generation method according to any one of claims 1 to 11.