Offline web subtitle editor

Abstract

With the advancement of the technology throughout the years, people are more reliant and inclined to technological devices for a more efficient and effective job completion. Millions of people surf the Internet to upload and view videos uploaded by people around the world using YouTube. Some of these videos consist of captions, enabling a wider range of people to understand videos of different language. In the case of captions, they are often being manually transcribed. During the transcribing process, transcribers listen to the audio of the videos and transcribe the voices into text. The process is very time-consuming and ineffective since transcribers are required to spend at least twice the amount of the length of the videos. With the development of speech recognition technology, captions are created automatically, and are available in different languages. These automatic captions are produced by machine learning algorithms, hence the accuracy may vary depending on the quality of videos. Therefore, transcribers are still needed to reinspect the quality of the captions to ensure that the transcript are of better quality. Thereafter, an analysis is done on the current different technologies of transcribing videos and audios, evaluating the advantages and disadvantages of existing tools, as well as integrating the techniques into the Offline Web Subtitle Editor.