首页> 外国专利> SYSTEM (EMBODIMENTS) FOR HARMONIOUSLY COMBINING VIDEO FILES AND AUDIO FILES AND CORRESPONDING METHOD

SYSTEM (EMBODIMENTS) FOR HARMONIOUSLY COMBINING VIDEO FILES AND AUDIO FILES AND CORRESPONDING METHOD

机译:系统(实施例)和谐地组合视频文件和音频文件和相应方法

摘要

The invention relates to computer systems, and more particularly to systems that support the processing of large sets of data using artificial intelligence. Proposed are embodiments of a system for harmoniously combining video files and audio files, comprising a server (100) and a user computer device (101), said server containing an intelligent system that comprises an artificial intelligence component with means for learning one or more machine learning algorithms and for analysing data in order to harmoniously combine video files and audio files, said intelligent system including: modules for collecting data and performing analytics for the training and operation of machine learning and data analysis models; analysis modules; modules for recommending audio parameters and video parameters; modules for searching for audio files and video files; modules for generating audio files and video files; and synchronization modules, wherein the video parameters are the following characteristics of a video file: objects, actions, mood of a video, activity and peaks, change in the lighting of a shot, change of colours, change of scene, speed of movement of the background relative to the foreground in a video file, sequence of shots, and video file metadata, and the audio parameters are the following parameters of an audio file: style, tempo, energy, activity and peaks, mood, acoustic quality, rhythmicality, instrumentation of the music, amount of sounds and noises, digital audio signal, and audio file metadata. Also proposed is a method for harmoniously combining video files and audio files, which is carried out using the claimed system.
机译:本发明涉及计算机系统,更具体地涉及支持使用人工智能的大集数据的系统。提出的是用于和谐地组合视频文件和音频文件的系统的实施例,包括服务器(100)和用户计算机设备(101),所述服务器包括智能系统,所述服务器包括具有用于学习一个或多个机器的手段的人工智能分量学习算法和分析数据以和谐地组合视频文件和音频文件,所述智能系统包括:用于收集数据和执行机器学习和数据分析模型的培训和操作的分析的模块;分析模块;用于推荐音频参数和视频参数的模块;搜索音频文件和视频文件的模块;用于生成音频文件和视频文件的模块;和同步模块,其中视频参数是视频文件的以下特征:对象,动作,视频,活动和峰值的情绪,变化在照明的镜头,变化的颜色,变化场景,运动速度背景技术相对于视频文件中的前景,拍摄序列和视频文件元数据,以及音频参数是音频文件的以下参数:风格,节奏,能量,活动和峰值,心情,声学质量,节奏性,音乐仪器,声音和噪声量,数字音频信号和音频文件元数据。还提出了一种和谐地组合视频文件和音频文件的方法,其使用所要求保护的系统执行。

著录项

  • 公开/公告号WO2022005442A1

    专利类型

  • 公开/公告日2022-01-06

    原文格式PDF

  • 申请/专利权人 PONOCHEVNYI NAZAR YURIEVYCH;

    申请/专利号WO2020UA00076

  • 发明设计人 PONOCHEVNYI NAZAR YURIEVYCH;

    申请日2020-08-04

  • 分类号H04N21/20;H04N21/234;H04N21/258;H04N21/2747;H04N21/4147;H04N21/80;G06F16/783;G06F16/61;G06F16/63;G06F16/65;G06F16/68;G06N3/02;G06N5/04;

  • 国家 UA

  • 入库时间 2022-08-24 23:16:31

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号