The present invention is the combination of image recognition and voice recognition technology uses the context at each scene in Audi's V extracts, this index (index) screen is by characterizing live metamessage and accurately by the way that user is by these Audi V and the easy search system for searching for enabling based on context and it is related to a search method using Audi of identical V V Audi. By supplier, it includes that sound provides that context-V audio retrievals system removes video according to the present invention, with the one or more folders of production, the characteristic of time, the expression of one characteristic, dialogue, at least one keyword, comprising: generate a context, its video information that folder is represented with the clip information of clip information, it provides folder, corresponding with a user query with clip information or context.
展开▼
