Retrieval System For Lecture Videos Based On Android

With the development of multimedia technology, distance education has been vigorously popularized. People has been used to obtain knowledge from the network video tutorials. But in the face of the rapid growth of lecture video resources, how to search the videos in our interest from the large amounts of video information is very important. Users always just want to give a keyword as query, then the system will be able to return the corresponding video list quickly and accurately. And with the advent of the mobile Internet, smart phone will become the indispensable equipment for people in the future. We should also realize that the retrieval for lecture videos in mobile application is missing, which has become a problem that we can not ignore. According to the above problems, the purpose of this paper is to achieve a lecture video retrieval system in the Android smart phone.In this paper, research is as follows:1. Overview of the current mainstream video retrieval technology. Research the situation of lecture videos in domestic and foreign, and analyze the disadvantages of existing methods, then put forward the idea of this paper:Turning the audio video into text with speech recognition engine Sphinx, and using the text to realize search function by Lucene.2. The default sorting of Lucene retrieval is according to the documents score. This paper introduced the user rating mechanism, with which system can combine feedback ratings with the results and return the results in diversity, so that we can meet the different needs of users.3. In this paper, the ultimate goal is to achieve an Android application. This application communicate with the server, and allow users to acquire new knowledge more convenient and free. The application can realize the function of logining、video list、video retrieval and video playback,etc.