RealPlayer Needed

You must install the RealPlayer plugin to use the Lecture Browser.

Welcome to the SLS Lecture Browser

The Lecture Browser is a web interface to video recordings of lectures and seminars that have been indexed using automatic speech recognition technology. You can search for topics, much like a regular web search engine. If any results look relevant, you can play the video starting at the relevant point and see the synchronized transcript.

• In order to use the interface, you will need to be able to play RealPlayer videos in your browser. The interface should work in Firefox and Internet Explorer, and may also work with other browsers.
• To use the interface, type in search terms (e.g., saxophone) and/or select a category from the Categories menu and then click on the search button.
• All lectures that contain words matching the search terms and category constraints will be shown. Segments of lectures with hits will be highlighted with a yellow oval.
• The first segment of the first lecture will be automatically expanded to show all individual "sentences" that matched the search terms. Different segments can be expanded by clicking on the button or contracted by clicking on the button under the lecture segment.
• Clicking on the button of any lecture segment or on any sentence words will start the video playing at that point. The automatically generated transcript will display synchronously with the video.
• You may also use the RealPlayer control buttons to pause or play the video, or drag the scroll bar to start the video at a different time.
• To provide more constraint to the search, enclose terms in ""'s that should occur together (e.g., "solar system"). Terms that should occur in the same lecture segment can be specified by AND (e.g., wine AND glass).

Spoken Lecture Processing

The Lecture Browser is part of a research project on spoken lecture processing that is being undertaken by researchers in the Spoken Language Systems Group at the MIT Computer Science and Artificial Intelligence Laboratory. Their goal is to enable natural speech-based interaction between humans and machines.

This project explores ways to enable educators and students to more effectively disseminate audio and video recordings of academic lecture material. To do this, we are developing technologies such as automatic speech recognition and language processing to help transcribe, annotate, structure, and even summarize audio-visual materials to help people search and explore these kind of data more easily. Our particular focus has been on recorded lectures that are being made available via initiatives such as MIT OpenCourseWare and MITWorld, in order to improve their accessibility to students or anyone interesting in learning from these educational materials.

The results of our research are being showcased in two different ways. In addition to the lecture browser shown here, we are also developing a a web-based spoken lecture processing server that allows users to upload audio files for automatic transcription and indexing. To help the speech recognizer, users can provide their own supplemental text files, such as journal articles, book chapters, etc., which can be used to adapt the language model and vocabulary of the system.

If you have any feedback you'd like to send us, send us email.

Support for this research was provided by the MIT/Microsoft iCampus Alliance for Educational Technology and by the National Science Foundation.