AUTOMATIC SPEECH RECOGNITION SYSTEM FOR LECTURE VIDEOS

The increase of video lecture data on World Wide Web is rapid therefore an efficient method of data retrieval is needed. So the system providing a method for data retrieval from the lecture video is implemented which will extract the speech data. Automatic video segmentation is applied first. Automatic Speech Recognition (ASR) applied on the lecture video tracks that are the speech contents given by lecturer during the video. The extracted text data is saved in the form of templates for future reference. The 80.97% of accuracy in recognising the letters from lecture videos is achieved and the extracted information is saved so that the quality of learning is improved.